{ "domain": "science.transductive.science", "generated_at": "2026-10-14T20:54:13.190818+00:00", "dossiers": [ { "group": "Foundations", "title": "Scientific protocol", "problem": "A written sequence is treated as the operative object; implementation drift becomes visible only after execution.", "repair": "Make protocols executable contracts with event logs, data schemas, instrument calibration and predeclared deviations.", "probe": "Run one protocol in two independent laboratories and diff every causal step.", "slug": "scientific-protocol", "number": 1, "source": { "requested_title": "Scientific protocol", "canonical_title": "Scientific protocol", "pageid": 9190286, "revision_id": 1321308276, "revision_timestamp": "2025-11-09T21:24:01Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_protocol", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1321308276", "intro": "In natural and social science research, a protocol is most commonly a predefined procedural method in the design and implementation of an experiment. Protocols are written whenever it is desirable to standardize a laboratory method to ensure successful replication of results by others in the same laboratory or by other laboratories. Additionally, and by extension, protocols have the advantage of facilitating the assessment of experimental results through peer review. In addition to detailed procedures, equipment, and instruments, protocols will also contain study objectives, reasoning for experimental design, reasoning for chosen sample sizes, safety precautions, and how results were calculated and reported, including statistical analysis and any rules for predefining and documenting excluded data to avoid bias.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489402+00:00" } }, { "group": "Foundations", "title": "Scientific method", "problem": "A human-scale hypothesis–experiment narrative poorly represents high-dimensional, parallel search.", "repair": "Track interacting populations of models, interventions and countermodels under explicit evidence budgets.", "probe": "Compare sequential and parallel model-search strategies on held-out scientific problems.", "slug": "scientific-method", "number": 2, "source": { "requested_title": "Scientific method", "canonical_title": "Scientific method", "pageid": 26833, "revision_id": 1376633181, "revision_timestamp": "2026-09-25T09:22:33Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_method", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376633181", "intro": "The scientific method is an empirical method for acquiring knowledge through careful observation, rigorous skepticism, hypothesis testing, and experimental validation. Developed from ancient and medieval practices, it acknowledges that cognitive assumptions can distort the interpretation of the observation. The scientific method has characterized science since at least the 17th century. Scientific inquiry includes creating a testable hypothesis through inductive reasoning, testing it through experiments and statistical analysis, and adjusting or discarding the hypothesis based on the results.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489424+00:00" } }, { "group": "Foundations", "title": "Philosophy of science", "problem": "Debates over method are often disconnected from measurable performance of actual inquiry.", "repair": "Turn competing philosophical commitments into testable policies for evidence management and discovery.", "probe": "Benchmark alternative methodological policies without assuming one doctrine is universal.", "slug": "philosophy-of-science", "number": 3, "source": { "requested_title": "Philosophy of science", "canonical_title": "Philosophy of science", "pageid": 37010, "revision_id": 1376331797, "revision_timestamp": "2026-09-23T14:13:41Z", "source_url": "https://en.wikipedia.org/wiki/Philosophy_of_science", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376331797", "intro": "Philosophy of science (also theory of science) is the branch of philosophy concerned with the foundations, methods, and implications of science. Amongst its central questions are the difference between science and non-science, the reliability of scientific theories, and the ultimate purpose and meaning of science as a human endeavour. Philosophy of science focuses on metaphysical, epistemic and semantic aspects of scientific practice, and overlaps with metaphysics, ontology, logic, and epistemology, for example, when it explores the relationship between science and the concept of truth. Philosophy of science is both a theoretical and empirical discipline, relying on philosophical theorising as well as meta-studies of scientific practice.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489430+00:00" } }, { "group": "Foundations", "title": "Epistemology", "problem": "Justification is frequently discussed without a machine-readable trail from proposition to observation.", "repair": "Represent reasons, counterarguments, uncertainty and defeaters as first-class claim dependencies.", "probe": "Retract one premise and measure whether all dependent claims are correctly flagged.", "slug": "epistemology", "number": 4, "source": { "requested_title": "Epistemology", "canonical_title": "Epistemology", "pageid": 9247, "revision_id": 1376802426, "revision_timestamp": "2026-09-26T08:31:56Z", "source_url": "https://en.wikipedia.org/wiki/Epistemology", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376802426", "intro": "Epistemology is the branch of philosophy that examines the nature, origin, and limits of knowledge. Also called the theory of knowledge, it explores different types of knowledge, such as propositional knowledge about facts, practical knowledge in the form of skills, and knowledge by acquaintance as a familiarity through experience. Epistemologists study the concepts of belief, truth, and justification to understand the nature of knowledge. To discover how knowledge arises, they investigate sources of justification, such as perception, introspection, memory, reason, and testimony.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489435+00:00" } }, { "group": "Foundations", "title": "Falsifiability", "problem": "The ability to imagine refutation does not rank competing explanations or localize failed assumptions.", "repair": "Generate interventions optimized to discriminate theory failure from auxiliary assumptions and instrument failure.", "probe": "Use adversarial synthetic worlds with several plausible failure modes.", "slug": "falsifiability", "number": 5, "source": { "requested_title": "Falsifiability", "canonical_title": "Falsifiability", "pageid": 11283, "revision_id": 1380088035, "revision_timestamp": "2026-10-13T18:30:30Z", "source_url": "https://en.wikipedia.org/wiki/Falsifiability", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1380088035", "intro": "Falsifiability is a standard of evaluation of scientific statements, including theories and hypotheses. A statement is falsifiable if it belongs to a language or logical structure capable of describing an empirical observation that contradicts it.\nIn the case of a theory, falsifiability requires that, given an initial condition, the theory must theoretically exclude some observations, that is, it must make formal predictions. It was introduced by the philosopher of science Karl Popper in his book The Logic of Scientific Discovery (1934).", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489438+00:00" } }, { "group": "Foundations", "title": "Problem of induction", "problem": "Past success cannot deductively guarantee future validity or extrapolation outside observed regimes.", "repair": "Require calibrated predictive distributions and systematic out-of-distribution challenge sets.", "probe": "Test models on deliberately shifted populations and report overconfidence.", "slug": "problem-of-induction", "number": 6, "source": { "requested_title": "Problem of induction", "canonical_title": "Problem of induction", "pageid": 177456, "revision_id": 1371948724, "revision_timestamp": "2026-08-29T14:47:00Z", "source_url": "https://en.wikipedia.org/wiki/Problem_of_induction", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1371948724", "intro": "The problem of induction is a philosophical problem that questions the rationality of predictions about unobserved things based on previous observations. These inferences from the observed to the unobserved are known as \"inductive inferences\". David Hume, who first formulated the problem in 1739, argued that there is no non-circular way to justify inductive inferences, while he acknowledged that everyone does and must make such inferences.\nThe traditional inductivist view is that all claimed empirical laws, either in everyday life or through the scientific method, can be justified through some form of reasoning.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489442+00:00" } }, { "group": "Foundations", "title": "Duhem–Quine thesis", "problem": "The tested statement is entangled with assumptions about apparatus, background theory and initial conditions.", "repair": "Build dependency-level diagnostic models that identify which assumption changes explain conflicting evidence.", "probe": "Insert hidden sensor faults and test whether diagnosis avoids falsely rejecting the main theory.", "slug": "duhemquine-thesis", "number": 7, "source": { "requested_title": "Duhem–Quine thesis", "canonical_title": "Duhem–Quine thesis", "pageid": 3130055, "revision_id": 1361617395, "revision_timestamp": "2026-06-29T02:10:05Z", "source_url": "https://en.wikipedia.org/wiki/Duhem%E2%80%93Quine_thesis", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1361617395", "intro": "In philosophy of science, the Duhem–Quine thesis, also called the Duhem–Quine problem, says that unambiguous falsifications of a scientific hypothesis are impossible, because an empirical test of the hypothesis requires one or more background assumptions. Rather than disproving the main hypothesis, the blame can be placed on one of the background beliefs or \"auxiliary\" hypotheses. It is named after French theoretical physicist Pierre Duhem and American logician Willard Van Orman Quine, who wrote about similar concepts. \nIn recent decades, the set of associated assumptions supporting a thesis sometimes is called a bundle of hypotheses, i.e.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489468+00:00" } }, { "group": "Foundations", "title": "Paradigm shift", "problem": "Disciplinary paradigms stabilize useful methods but can suppress representation-level alternatives.", "repair": "Run competing ontologies in parallel and compare their consequences on common experimental interfaces.", "probe": "Measure whether alternative representations discover variables missed by incumbent models.", "slug": "paradigm-shift", "number": 8, "source": { "requested_title": "Paradigm shift", "canonical_title": "Paradigm shift", "pageid": 25081, "revision_id": 1360663769, "revision_timestamp": "2026-06-22T21:17:18Z", "source_url": "https://en.wikipedia.org/wiki/Paradigm_shift", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1360663769", "intro": "A paradigm shift is a fundamental change in the basic concepts and experimental practices of a scientific discipline. It is a concept in the philosophy of science that was introduced and brought into the common lexicon by the American physicist and philosopher Thomas Kuhn. Even though Kuhn restricted the use of the term to the natural sciences, the concept of a paradigm shift has also been used in numerous non-scientific contexts to describe a profound change in a fundamental model or perception of events.\nKuhn presented his notion of a paradigm shift in his influential book The Structure of Scientific Revolutions (1962).", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489473+00:00" } }, { "group": "Foundations", "title": "Underdetermination", "problem": "Several theories can fit the same observations; agreement with history alone is insufficient.", "repair": "Maintain explicit equivalence classes of observationally compatible models and seek separating experiments.", "probe": "Generate cases with identical training fit but divergent intervention predictions.", "slug": "underdetermination", "number": 9, "source": { "requested_title": "Underdetermination", "canonical_title": "Underdetermination", "pageid": 1678145, "revision_id": 1362414613, "revision_timestamp": "2026-07-03T20:23:43Z", "source_url": "https://en.wikipedia.org/wiki/Underdetermination", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1362414613", "intro": "In the philosophy of science, underdetermination or the underdetermination of theory by data (sometimes abbreviated UTD) is the idea that evidence available to us at a given time may be insufficient to determine what beliefs we should hold in response to it. The underdetermination thesis states that all evidence necessarily underdetermines any scientific theory.\nUnderdetermination exists when available evidence is insufficient to identify which belief one should hold about that evidence. For example, if all that was known was that exactly $10 were spent on apples and oranges, and that apples cost $1 and oranges $2, then one would know enough to eliminate some possibilities (e.g., 6 oranges could not have been purchased), but one would not have enough evidence to know which specific combination of apples and oranges were purchased.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489476+00:00" } }, { "group": "Foundations", "title": "Scientific realism", "problem": "A successful predictive model can be confused with a settled claim about underlying entities.", "repair": "Keep predictive adequacy, causal interpretation and ontological commitment as separately supported properties.", "probe": "Require each proposed entity to earn independent observable constraints.", "slug": "scientific-realism", "number": 10, "source": { "requested_title": "Scientific realism", "canonical_title": "Scientific realism", "pageid": 270804, "revision_id": 1370998013, "revision_timestamp": "2026-08-24T02:36:58Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_realism", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1370998013", "intro": "Scientific realism is the philosophical view that the universe described by science (including both observable and unobservable aspects) exists independently of perception, and that verified scientific theories are at least approximately true descriptions of what is real. Scientific realists typically assert that science, when successful, uncovers true (or approximately true) knowledge about nature, including aspects of reality that are not directly observable.\nWithin philosophy of science, this view is often an answer to the question \"how is the success of science to be explained?\" The discussion on the success of science in this context centers primarily on the status of unobservable entities apparently talked about by scientific theories. Generally, those who are scientific realists assert that one can make valid claims about unobservables (viz., that they have the same ontological status) as observables, as opposed to instrumentalism.", "verified": true, "retrieved_at": "2026-10-14T20:47:00.489480+00:00" } }, { "group": "Foundations", "title": "Scientific model", "problem": "Static models hide version history, parameter changes and domains where predictions break.", "repair": "Maintain executable, versioned model cards with dependencies, domain checks and failure maps.", "probe": "Replay predictions across historical datasets and controlled domain shifts.", "slug": "scientific-model", "number": 11, "source": { "requested_title": "Scientific model", "canonical_title": "Scientific modelling", "pageid": 3224795, "revision_id": 1371659107, "revision_timestamp": "2026-08-27T18:11:48Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_modelling", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1371659107", "intro": "Scientific modelling is an activity that produces models representing empirical objects, phenomena, and physical processes, to make a particular part or feature of the world easier to understand, define, quantify, visualize, or simulate. It requires selecting and identifying relevant aspects of a situation in the real world and then developing a model to replicate a system with those features. Different types of models may be used for different purposes, such as conceptual models to better understand, operational models to operationalize, mathematical models to quantify, computational models to simulate, and graphical models to visualize the subject. \nModelling is an essential and inseparable part of many scientific disciplines, each of which has its own ideas about specific types of modelling.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411658+00:00" } }, { "group": "Foundations", "title": "Occam's razor", "problem": "Simplicity can become aesthetic authority rather than a calibrated trade-off.", "repair": "Compare model complexity with predictive adequacy, interpretability and robustness using explicit criteria.", "probe": "Evaluate simple and complex models on withheld regimes where each may fail.", "slug": "occam-s-razor", "number": 12, "source": { "requested_title": "Occam's razor", "canonical_title": "Occam's razor", "pageid": 36797, "revision_id": 1379603268, "revision_timestamp": "2026-10-10T22:52:05Z", "source_url": "https://en.wikipedia.org/wiki/Occam%27s_razor", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379603268", "intro": "In philosophy, Occam's razor (also spelled Ockham's razor or Ocham's razor; Latin: novacula Occami) is the problem-solving principle that recommends searching for explanations constructed with the smallest possible set of elements. It is also known as the principle of parsimony or the law of parsimony (Latin: lex parsimoniae). Attributed to William of Ockham, a 14th-century English philosopher and theologian, it is frequently cited as Entia non sunt multiplicanda praeter necessitatem, which translates as \"Entities must not be multiplied beyond necessity\", although Occam never used those words. Popularly, the principle is sometimes paraphrased as \"of two competing theories, the simpler explanation of an entity is to be preferred\".", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411686+00:00" } }, { "group": "Inference", "title": "Bayesian inference", "problem": "Posterior certainty is conditional on model classes and priors that are often left unchallenged.", "repair": "Audit priors, likelihoods and missing-model risk while comparing alternative model families.", "probe": "Use simulation-based calibration and sensitivity analyses on intentionally misspecified models.", "slug": "bayesian-inference", "number": 13, "source": { "requested_title": "Bayesian inference", "canonical_title": "Bayesian inference", "pageid": 49571, "revision_id": 1379760658, "revision_timestamp": "2026-10-11T20:20:55Z", "source_url": "https://en.wikipedia.org/wiki/Bayesian_inference", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379760658", "intro": "Bayesian inference ( BAY-zee-ən or BAY-zhən) is a method of statistical inference in which Bayes' theorem is used to calculate a probability of a hypothesis, given prior evidence, and update it as more information becomes available. Fundamentally, Bayesian inference uses a prior distribution to estimate posterior probabilities. Bayesian inference is an important technique in statistics, and especially in mathematical statistics. Bayesian updating is particularly important in the dynamic analysis of a sequence of data.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411691+00:00" } }, { "group": "Inference", "title": "Statistical hypothesis test", "problem": "A thresholded decision compresses uncertainty, design choices and effect sizes into a ritual verdict.", "repair": "Return effect estimates, decision losses, selection adjustments and predictive checks beside test results.", "probe": "Test whether consequential decisions improve beyond a binary significance rule.", "slug": "statistical-hypothesis-test", "number": 14, "source": { "requested_title": "Statistical hypothesis test", "canonical_title": "Statistical hypothesis test", "pageid": 30284, "revision_id": 1379362199, "revision_timestamp": "2026-10-09T14:18:31Z", "source_url": "https://en.wikipedia.org/wiki/Statistical_hypothesis_test", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379362199", "intro": "A statistical hypothesis test is a method of statistical inference used to decide whether the data provides sufficient evidence to reject a particular hypothesis. A statistical hypothesis test typically involves a calculation of a test statistic. Then a decision is made, either by comparing the test statistic to a critical value or equivalently by evaluating a p-value computed from the test statistic. Roughly 100 specialized statistical tests are in use.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411695+00:00" } }, { "group": "Inference", "title": "Null hypothesis", "problem": "A point null is sometimes operationally convenient but mistaken for a meaningful physical absence.", "repair": "Make practically relevant effect intervals and competing causal claims explicit.", "probe": "Compare equivalence tests and estimation against point-null-only reporting.", "slug": "null-hypothesis", "number": 15, "source": { "requested_title": "Null hypothesis", "canonical_title": "Statistical hypothesis test", "pageid": 30284, "revision_id": 1379362199, "revision_timestamp": "2026-10-09T14:18:31Z", "source_url": "https://en.wikipedia.org/wiki/Statistical_hypothesis_test", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379362199", "intro": "A statistical hypothesis test is a method of statistical inference used to decide whether the data provides sufficient evidence to reject a particular hypothesis. A statistical hypothesis test typically involves a calculation of a test statistic. Then a decision is made, either by comparing the test statistic to a critical value or equivalently by evaluating a p-value computed from the test statistic. Roughly 100 specialized statistical tests are in use.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411700+00:00" } }, { "group": "Inference", "title": "Statistical significance", "problem": "A p-value is often reinterpreted as the probability a hypothesis is true.", "repair": "Generate inference reports that distinguish tail-area probabilities, effect magnitude and model assumptions.", "probe": "Automatically detect known p-value interpretation mistakes in published abstracts.", "slug": "statistical-significance", "number": 16, "source": { "requested_title": "Statistical significance", "canonical_title": "Statistical significance", "pageid": 160995, "revision_id": 1361974577, "revision_timestamp": "2026-07-01T05:07:56Z", "source_url": "https://en.wikipedia.org/wiki/Statistical_significance", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1361974577", "intro": "In statistical hypothesis testing, a result has statistical significance when a result at least as extreme would be very infrequent if the null hypothesis were true. More precisely, a study's defined significance level, denoted by \n \n \n \n α\n \n \n {\\displaystyle \\alpha }\n \n, is the probability of the study rejecting the null hypothesis, given that the null hypothesis is true; and the p-value of a result, \n \n \n \n p\n \n \n {\\displaystyle p}\n \n, is the probability of obtaining a result at least as extreme, given that the null hypothesis is true. The result is said to be statistically significant, by the standards of the study, when \n \n \n \n p\n ≤\n α\n \n \n {\\displaystyle p\\leq \\alpha }\n \n. The significance level for a study is chosen before data collection, and is typically set to 5% or much lower—depending on the field of study.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411704+00:00" } }, { "group": "Inference", "title": "Statistical power", "problem": "Underpowered experiments generate unstable estimates and encourage selective attention to large effects.", "repair": "Optimize sampling plans for precision and actionable outcomes under fixed resource constraints.", "probe": "Compare adaptive designs against conventional sample-size recipes.", "slug": "statistical-power", "number": 17, "source": { "requested_title": "Statistical power", "canonical_title": "Power (statistics)", "pageid": 238695, "revision_id": 1378995428, "revision_timestamp": "2026-10-07T07:40:16Z", "source_url": "https://en.wikipedia.org/wiki/Power_(statistics)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1378995428", "intro": "In frequentist statistics, power is the probability of detecting an effect (i.e. rejecting the null hypothesis) given that some prespecified effect actually exists using a given test in a given context. In typical use, it is a function of the specific test that is used (including the choice of test statistic and significance level), the sample size (more data tends to provide more power), and the effect size (effects or correlations that are large relative to the variability of the data tend to provide more power).\nMore formally, in the case of a simple hypothesis test with two hypotheses, the power of the test is the probability that the test correctly rejects the null hypothesis (\n \n \n \n \n H\n \n 0\n \n \n \n \n {\\displaystyle H_{0}}\n \n) when the alternative hypothesis (\n \n \n \n \n H\n \n 1\n \n \n \n \n {\\displaystyle H_{1}}\n \n) is true.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411707+00:00" } }, { "group": "Inference", "title": "Multiple comparisons problem", "problem": "Hidden flexibility in the number of questions tested inflates apparently impressive findings.", "repair": "Log every analysis path and apply selection-aware inference or honest independent validation.", "probe": "Audit large search pipelines with deliberately planted null signals.", "slug": "multiple-comparisons-problem", "number": 18, "source": { "requested_title": "Multiple comparisons problem", "canonical_title": "Multiple comparisons problem", "pageid": 9444220, "revision_id": 1376255469, "revision_timestamp": "2026-09-23T01:59:46Z", "source_url": "https://en.wikipedia.org/wiki/Multiple_comparisons_problem", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376255469", "intro": "Multiple comparisons, multiplicity or multiple testing problem occurs when many statistical tests are performed on the same dataset. Each test has its own chance of a Type I error (false positive), so the overall probability of making at least one false positive increases as the number of tests grows. In statistics, this occurs when one simultaneously considers a set of statistical inferences or estimates a subset of selected parameters based on observed values.\nThe probability of false positives is measured through the family-wise error rate (FWER).", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411711+00:00" } }, { "group": "Inference", "title": "False discovery rate", "problem": "Global error controls can be applied without confronting dependence or shifting hypothesis families.", "repair": "Track test dependence and streaming decisions with auditable false-discovery budgets.", "probe": "Stress-test correlated tests under changing hypothesis streams.", "slug": "false-discovery-rate", "number": 19, "source": { "requested_title": "False discovery rate", "canonical_title": "False discovery rate", "pageid": 4574271, "revision_id": 1374927830, "revision_timestamp": "2026-09-14T22:08:05Z", "source_url": "https://en.wikipedia.org/wiki/False_discovery_rate", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1374927830", "intro": "In statistics, the false discovery rate (FDR) is a method of conceptualizing the rate of type I errors in null hypothesis testing when conducting multiple comparisons. FDR-controlling procedures are designed to control the FDR, which is the expected proportion of \"discoveries\" (rejected null hypotheses) that are false (incorrect rejections of the null). Equivalently, the FDR is the expected ratio of the number of false positive classifications (false discoveries) to the total number of positive classifications (rejections of the null). The total number of rejections of the null include both the number of false positives (FP) and true positives (TP).", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411714+00:00" } }, { "group": "Inference", "title": "Confidence interval", "problem": "An interval may be given a probability interpretation it does not support and may omit selection effects.", "repair": "Display its construction assumptions, repeated-sampling meaning and sensitivity to alternative procedures.", "probe": "Check empirical coverage after model choice and post-selection adjustments.", "slug": "confidence-interval", "number": 20, "source": { "requested_title": "Confidence interval", "canonical_title": "Confidence interval", "pageid": 280911, "revision_id": 1376255302, "revision_timestamp": "2026-09-23T01:58:37Z", "source_url": "https://en.wikipedia.org/wiki/Confidence_interval", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376255302", "intro": "In frequentist inference, a confidence interval (CI) determines lower and upper bounds likely to contain (in repeated sampling) the true value of an unknown statistical parameter, such as a population mean. Rather than reporting a single point estimate (e.g. \"the average screen time is 3 hours per day\"), a confidence interval provides an interval, such as from 2 hours to 4 hours.\nA confidence interval is based on a pre-defined coverage probability, called confidence level, typically 95%.", "verified": true, "retrieved_at": "2026-10-14T20:47:01.411718+00:00" } }, { "group": "Inference", "title": "Meta-analysis", "problem": "Combining available studies can precisely estimate a biased published subset.", "repair": "Construct study-inception registries and selection models including missing or unpublished results.", "probe": "Compare pooled estimates with and without identified missing studies.", "slug": "meta-analysis", "number": 21, "source": { "requested_title": "Meta-analysis", "canonical_title": "Meta-analysis", "pageid": 62329, "revision_id": 1379623909, "revision_timestamp": "2026-10-11T01:10:53Z", "source_url": "https://en.wikipedia.org/wiki/Meta-analysis", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379623909", "intro": "Meta-analysis is a method of synthesis of quantitative data from multiple independent studies addressing a common research question. An important part of this method involves computing a combined effect size across all of the studies. As such, this statistical approach involves extracting effect sizes and variance measures from various studies. By combining these effect sizes the statistical power is improved and can resolve uncertainties or discrepancies found in individual studies.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412539+00:00" } }, { "group": "Inference", "title": "Regression toward the mean", "problem": "Extreme observations can generate spurious improvement or deterioration narratives.", "repair": "Generate matched control trajectories and counterfactual forecasts before attributing effects.", "probe": "Challenge before–after designs with regression artifacts.", "slug": "regression-toward-the-mean", "number": 22, "source": { "requested_title": "Regression toward the mean", "canonical_title": "Regression toward the mean", "pageid": 186028, "revision_id": 1368360447, "revision_timestamp": "2026-08-08T16:22:25Z", "source_url": "https://en.wikipedia.org/wiki/Regression_toward_the_mean", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1368360447", "intro": "In statistics, regression toward the mean (also called regression to the mean, reversion to the mean, and reversion to mediocrity) is the phenomenon where if one sample of a random variable is extreme, the next sampling of the same random variable is likely to be closer to its mean. Furthermore, when many random variables are sampled and the most extreme results are intentionally picked out, it refers to the fact that (in many cases) a second sampling of these picked-out variables will result in \"less extreme\" results, closer to the initial mean of all of the variables.\nMathematically, the strength of this \"regression\" effect is dependent on whether or not all of the random variables are drawn from the same distribution, or if there are genuine differences in the underlying distributions for each random variable. In the first case, the \"regression\" effect is statistically likely to occur, but in the second case, it may occur less strongly or not at all.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412597+00:00" } }, { "group": "Bias", "title": "Data dredging", "problem": "Exploratory searches are routinely mistaken for tests whose evidential threshold was fixed beforehand.", "repair": "Permit unlimited exploration but quarantine confirmatory evidence and record dataset exposure.", "probe": "Have agents explore thousands of predictors then verify discoveries on untouched data.", "slug": "data-dredging", "number": 23, "source": { "requested_title": "Data dredging", "canonical_title": "Data dredging", "pageid": 1311951, "revision_id": 1377936004, "revision_timestamp": "2026-10-02T01:26:30Z", "source_url": "https://en.wikipedia.org/wiki/Data_dredging", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1377936004", "intro": "Data dredging, also known as data snooping or p-hacking, is the misuse of data analysis to find patterns in data that can be presented as statistically significant, thus dramatically increasing and understating the risk of false positives. This is done by performing many statistical tests on the data and only reporting those that come back with significant results. Thus data dredging is also often a misused or misapplied form of data mining.\nThe process of data dredging involves testing multiple hypotheses using a single data set by exhaustively searching—perhaps for combinations of variables that might show a correlation, and perhaps for groups of cases or observations that show differences in their mean or in their breakdown by some other variable.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412608+00:00" } }, { "group": "Bias", "title": "Publication bias", "problem": "The scientific record is filtered by publishability and statistical excitement.", "repair": "Register investigation inception and deposit all outcomes, including null and stopped work.", "probe": "Audit the difference between started trials and accessible results.", "slug": "publication-bias", "number": 24, "source": { "requested_title": "Publication bias", "canonical_title": "Publication bias", "pageid": 511115, "revision_id": 1368850151, "revision_timestamp": "2026-08-11T12:10:58Z", "source_url": "https://en.wikipedia.org/wiki/Publication_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1368850151", "intro": "In published academic research, publication bias occurs when the outcome of an experiment or research study biases the decision to publish or otherwise distribute it. Publishing only results that show a significant finding disturbs the balance of findings in favor of positive results. The study of publication bias is an important topic in metascience.\nDespite similar quality of execution and design, papers with statistically significant results are three times more likely to be published than those with null results.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412615+00:00" } }, { "group": "Bias", "title": "Confirmation bias", "problem": "Investigators preferentially see support for expected explanations.", "repair": "Assign independent adversarial agents to construct disconfirming tests and alternative interpretations.", "probe": "Blind reviewers to model identity and measure detection of planted contradictions.", "slug": "confirmation-bias", "number": 25, "source": { "requested_title": "Confirmation bias", "canonical_title": "Confirmation bias", "pageid": 59160, "revision_id": 1370294982, "revision_timestamp": "2026-08-20T07:18:49Z", "source_url": "https://en.wikipedia.org/wiki/Confirmation_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1370294982", "intro": "Confirmation bias (also confirmatory bias, myside bias, or congeniality bias) is the tendency to search for, interpret, favor and recall information in a way that confirms or supports one's prior beliefs, values, or decisions. People display this bias when they select information that supports their views, ignoring contrary information or when they interpret ambiguous evidence as supporting their existing attitudes. The effect is strongest for desired outcomes, emotionally charged issues and deeply entrenched beliefs. Confirmation bias often comes from automatic mental habits.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412621+00:00" } }, { "group": "Bias", "title": "Selection bias", "problem": "Who enters the dataset can determine the apparent result.", "repair": "Record inclusion mechanisms and compute sensitivity to unobserved sampling and missingness.", "probe": "Compare source and target populations with known selection distortions.", "slug": "selection-bias", "number": 26, "source": { "requested_title": "Selection bias", "canonical_title": "Selection bias", "pageid": 394392, "revision_id": 1375130988, "revision_timestamp": "2026-09-16T01:14:20Z", "source_url": "https://en.wikipedia.org/wiki/Selection_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375130988", "intro": "Selection bias is the bias introduced by the selection of individuals, groups, or data for analysis in such a way that the association between exposure and outcome among those selected for analysis differs from the association among those eligible. It typically occurs when researchers condition on a factor that is influenced both by the exposure and the outcome (or their causes), creating a false association between them. Selection bias encompasses several forms of bias, including differential loss-to-follow-up, incidence–prevalence bias, volunteer bias, healthy-worker bias, and nonresponse bias.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412627+00:00" } }, { "group": "Bias", "title": "Survivorship bias", "problem": "Only successful or visible cases inform a seemingly convincing causal story.", "repair": "Enumerate missing and failed cases as part of the evidence model.", "probe": "Introduce invisible failures and test whether aggregate conclusions remain valid.", "slug": "survivorship-bias", "number": 27, "source": { "requested_title": "Survivorship bias", "canonical_title": "Survivorship bias", "pageid": 1745325, "revision_id": 1379606971, "revision_timestamp": "2026-10-10T23:22:37Z", "source_url": "https://en.wikipedia.org/wiki/Survivorship_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379606971", "intro": "Survivorship bias or survivor bias is a statistical error which results from concentrating on entities that passed a selection process over those that did not. This can lead to incorrect conclusions because of incomplete data.\nSurvivorship bias may lead to overly optimistic beliefs because multiple failures are overlooked, such as when companies that no longer exist are excluded from analyses of financial performance. It can also lead to the false belief that the successes in a group have some special property, rather than just a coincidence, as in correlation \"proves\" causality.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412634+00:00" } }, { "group": "Bias", "title": "Sampling bias", "problem": "A large sample can be precise yet systematically unrepresentative.", "repair": "Link sampling plans to population models and estimate transport uncertainty.", "probe": "Use shifted population benchmarks to test generalization claims.", "slug": "sampling-bias", "number": 28, "source": { "requested_title": "Sampling bias", "canonical_title": "Sampling bias", "pageid": 17692, "revision_id": 1375044021, "revision_timestamp": "2026-09-15T14:47:23Z", "source_url": "https://en.wikipedia.org/wiki/Sampling_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375044021", "intro": "In statistics, sampling bias is a bias in which a sample is collected in such a way that some members of the intended population have a lower or higher sampling probability than others. It results in a biased sample of a population (or non-human factors) in which all individuals, or instances, were not equally likely to have been selected. If this is not accounted for, results can be erroneously attributed to the phenomenon under study rather than to the method of sampling.\nMedical sources sometimes refer to sampling bias as ascertainment bias.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412639+00:00" } }, { "group": "Bias", "title": "HARKing", "problem": "A post-hoc explanation is presented as if it were the prediction that motivated the work.", "repair": "Timestamp candidate hypotheses and distinguish discovered patterns from genuinely prospective tests.", "probe": "Compare declared predictions against immutable research traces.", "slug": "harking", "number": 29, "source": { "requested_title": "HARKing", "canonical_title": "HARKing", "pageid": 63211960, "revision_id": 1359619333, "revision_timestamp": "2026-06-16T07:22:03Z", "source_url": "https://en.wikipedia.org/wiki/HARKing", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1359619333", "intro": "HARKing (hypothesizing after the results are known) is an acronym coined by social psychologist Norbert Kerr that refers to the questionable research practice of \"presenting a post hoc hypothesis in the introduction of a research report as if it were an a priori hypothesis\". Hence, a key characteristic of HARKing is that post hoc hypothesizing is falsely portrayed as a priori hypothesizing. HARKing may occur when a researcher tests an a priori hypothesis but then omits that hypothesis from their research report after they find out the results of their test. Post hoc analysis or post hoc theorizing then may lead to a post hoc hypothesis.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412646+00:00" } }, { "group": "Bias", "title": "Pseudoreplication", "problem": "Dependent observations are counted as if they were independent experimental units.", "repair": "Infer hierarchical structure and enforce correct experimental-unit accounting.", "probe": "Evaluate nested, clustered datasets under naive and corrected inference.", "slug": "pseudoreplication", "number": 30, "source": { "requested_title": "Pseudoreplication", "canonical_title": "Pseudoreplication", "pageid": 10159743, "revision_id": 1367999065, "revision_timestamp": "2026-08-06T12:46:39Z", "source_url": "https://en.wikipedia.org/wiki/Pseudoreplication", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1367999065", "intro": "Pseudoreplication (sometimes unit of analysis error) has many definitions. Pseudoreplication was originally defined in 1984 by Stuart H. Hurlbert as the use of inferential statistics to test for treatment effects with data from experiments where either treatments are not replicated (though samples may be) or\nreplicates are not statistically independent. Subsequently, Millar and Anderson identified it as a special case of inadequate specification of random factors where both random and fixed factors are present. It is sometimes narrowly interpreted as an inflation of the number of samples or replicates which are not statistically independent.", "verified": true, "retrieved_at": "2026-10-14T20:47:02.412657+00:00" } }, { "group": "Bias", "title": "Cognitive bias", "problem": "Human heuristic errors are treated as individual defects despite being amplified by institutions.", "repair": "Monitor decision pipelines for systematic judgment failures, including correlated AI mistakes.", "probe": "Run blind evidence-assessment competitions across human and AI teams.", "slug": "cognitive-bias", "number": 31, "source": { "requested_title": "Cognitive bias", "canonical_title": "Cognitive bias", "pageid": 47278, "revision_id": 1379095929, "revision_timestamp": "2026-10-07T23:04:20Z", "source_url": "https://en.wikipedia.org/wiki/Cognitive_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379095929", "intro": "A cognitive bias is a systematic pattern of deviation from the norm or rationality in judgment. Individuals create their own \"subjective reality\" from their perception of the input. An individual's construction of reality, not the objective input, may dictate their behavior in the world. Thus, cognitive biases may sometimes lead to perceptual distortion, inaccurate judgment, illogical interpretation, and irrationality.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322705+00:00" } }, { "group": "Bias", "title": "Observer bias", "problem": "Instrument and coding choices can systematically shift observations before analysis begins.", "repair": "Record sensor lineage, calibration, annotator process and measurement uncertainty as causal variables.", "probe": "Inject calibration drift and label bias, then measure anomaly detection.", "slug": "observer-bias", "number": 32, "source": { "requested_title": "Observer bias", "canonical_title": "Observer bias", "pageid": 599870, "revision_id": 1353554158, "revision_timestamp": "2026-05-10T23:27:32Z", "source_url": "https://en.wikipedia.org/wiki/Observer_bias", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1353554158", "intro": "Observer bias is one of the types of detection bias and is defined as any kind of systematic divergence from accurate facts during observation and the recording of data and information in studies. The definition can be further expanded upon to include the systematic difference between what is observed due to variation in observers, and what the true value is.\nObserver bias is the tendency of observers to not see what is there, but instead to see what they expect or want to see. This is a common occurrence in the everyday lives of many and is a significant problem that is sometimes encountered in scientific research and studies.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322720+00:00" } }, { "group": "Reliability", "title": "Replication crisis", "problem": "Validation arrives late, after strong claims and reputational accumulation.", "repair": "Schedule independent replication as a default evidence stage tied to claim importance and uncertainty.", "probe": "Measure time from publication to detection of injected non-replicable results.", "slug": "replication-crisis", "number": 33, "source": { "requested_title": "Replication crisis", "canonical_title": "Replication crisis", "pageid": 44984325, "revision_id": 1379868267, "revision_timestamp": "2026-10-12T10:01:54Z", "source_url": "https://en.wikipedia.org/wiki/Replication_crisis", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379868267", "intro": "The replication crisis, also known as the reproducibility crisis or replicability crisis, is a widespread phenomenon in which published scientific results cannot be reproduced. Because the reproducibility of empirical results is the cornerstone of the scientific method, such failures undermine the credibility of theories and challenge substantial parts of scientific knowledge.\nPsychology and medicine have been focal points for replication efforts, with researchers systematically reexamining classic studies to verify their reliability and, when failures emerge, to identify the underlying causes. Data strongly indicates that other natural and social sciences are also affected.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322724+00:00" } }, { "group": "Reliability", "title": "Reproducibility", "problem": "Code and methods may be described but not executable outside the authors' environment.", "repair": "Archive runnable computational environments, input hashes, expected outputs and execution receipts.", "probe": "Replay published pipelines under clean environments on independent hosts.", "slug": "reproducibility", "number": 34, "source": { "requested_title": "Reproducibility", "canonical_title": "Reproducibility", "pageid": 47651, "revision_id": 1379031988, "revision_timestamp": "2026-10-07T14:07:16Z", "source_url": "https://en.wikipedia.org/wiki/Reproducibility", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379031988", "intro": "Reproducibility, closely related to replicability and repeatability, is a major principle underpinning the scientific method. For the findings of a study to be reproducible means that results obtained by an experiment or an observational study or in a statistical analysis of a data set should be achieved again with a high degree of reliability when the study is replicated. There are different kinds of replication but typically replication studies involve different researchers using the same methodology. Only after one or several such successful replications should a result be recognized as scientific knowledge.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322728+00:00" } }, { "group": "Reliability", "title": "Replication (statistics)", "problem": "Replication is treated as a new paper rather than a continuously linked evidence update.", "repair": "Attach independent tests and population-specific effect estimates to a persistent claim object.", "probe": "Compare claim confidence before and after multiple independent replications.", "slug": "replication-statistics", "number": 35, "source": { "requested_title": "Replication (statistics)", "canonical_title": "Replication (statistics)", "pageid": 10306262, "revision_id": 1360732077, "revision_timestamp": "2026-06-23T07:06:44Z", "source_url": "https://en.wikipedia.org/wiki/Replication_(statistics)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1360732077", "intro": "In engineering, science, and statistics, replication is the process of repeating a study or experiment under the same or similar conditions. It is a crucial step to test the original claim and confirm or reject the accuracy of results as well as for identifying and correcting the flaws in the original experiment. ASTM, in standard E1847, defines replication as \"... the repetition of the set of all the treatment combinations to be compared in an experiment.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322732+00:00" } }, { "group": "Reliability", "title": "Peer review", "problem": "A limited set of reviewers acts as a bottleneck and imperfect proxy for validity.", "repair": "Continuous adversarial audits with specialist humans at unresolved boundaries and independent reproducibility checks.", "probe": "Benchmark reviewers and agents on blinded corpora of seeded scientific mistakes.", "slug": "peer-review", "number": 36, "source": { "requested_title": "Peer review", "canonical_title": "Peer review", "pageid": 24116, "revision_id": 1379295044, "revision_timestamp": "2026-10-09T02:34:08Z", "source_url": "https://en.wikipedia.org/wiki/Peer_review", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379295044", "intro": "Peer review is the evaluation of work by one or more people with similar competencies as the producers of the work (peers). It functions as a form of self-regulation by qualified members of a profession within the relevant field. \nPeer review methods are used to maintain quality standards, improve performance, and provide credibility. In academia, scholarly peer review is typically used to determine an academic paper's suitability for publication.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322736+00:00" } }, { "group": "Reliability", "title": "Metascience", "problem": "Reform is studied in disconnected publications, often far from the research workflows being evaluated.", "repair": "Run controlled comparisons of scientific methods as part of the same execution infrastructure.", "probe": "Randomize projects across verification methods and observe downstream error correction.", "slug": "metascience", "number": 37, "source": { "requested_title": "Metascience", "canonical_title": "Metascience", "pageid": 53147698, "revision_id": 1373322115, "revision_timestamp": "2026-09-05T08:22:01Z", "source_url": "https://en.wikipedia.org/wiki/Metascience", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1373322115", "intro": "Metascience, also known as science of science, is the systematic study of science itself. It analyzes the practices, structures, and outcomes of scientific research, including research design, peer review, publication bias, replication, and incentive structures, with the aim of improving the reliability, transparency, and impact of scientific knowledge production. Emerging at the intersection of the philosophy of science, the sociology of science, scientometrics, and research methodology, it draws on empirical and quantitative approaches to evaluate how science is conducted, communicated, and assessed. \nThe field has gained prominence in the context of the replication crisis and broader debates around open science, research integrity, and evidence-based policy.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322739+00:00" } }, { "group": "Reliability", "title": "Open science", "problem": "Sharing artifacts does not ensure they are usable, correctly linked, or independently inspectable.", "repair": "Make open records machine-actionable with persistent identifiers, provenance and executable tests.", "probe": "Measure whether unrelated teams can independently rerun and extend a result.", "slug": "open-science", "number": 38, "source": { "requested_title": "Open science", "canonical_title": "Open science", "pageid": 6277878, "revision_id": 1379400605, "revision_timestamp": "2026-10-09T19:51:08Z", "source_url": "https://en.wikipedia.org/wiki/Open_science", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379400605", "intro": "Open science (also known as open research) is the movement to make scientific research, including publications, data, physical samples, software, and models, transparent and accessible to all levels of society through collaborative networks. This encompasses practices such as publishing open research, campaigning for open access, encouraging scientists to practice open-notebook science (such as openly sharing data and code and now open hardware to enable replication of physical experiments), broader dissemination and public engagement in science, and generally making it easier to publish, access, and communicate scientific knowledge.\nUsage of the term varies substantially across disciplines, with a notable prevalence in the STEM disciplines. The term 'open research' has gained currency as a broader alternative to 'open science,' encompassing the humanities and arts alongside traditional scientific disciplines.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322743+00:00" } }, { "group": "Reliability", "title": "Preregistration (science)", "problem": "Research intentions can be registered but later analytical decisions remain hard to reconstruct.", "repair": "Bind time-stamped commitments to executable statistical plans and append-only deviation records.", "probe": "Compare protocol deviations detectable by automation against manual review.", "slug": "preregistration-science", "number": 39, "source": { "requested_title": "Preregistration (science)", "canonical_title": "Preregistration (science)", "pageid": 59031980, "revision_id": 1372960129, "revision_timestamp": "2026-09-03T04:55:44Z", "source_url": "https://en.wikipedia.org/wiki/Preregistration_(science)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1372960129", "intro": "Preregistration is the practice of registering the hypotheses, methods, or analyses of a scientific study before it is conducted. Clinical trial registration is similar, although it may not require the registration of a study's analysis protocol. Finally, registered reports include the peer review and in principle acceptance of a study protocol prior to data collection.\nPreregistration has the goal to transparently evaluate the severity of hypothesis tests, and can have a number of secondary goals (which can also be achieved without preregistering ), including (a) facilitating and documenting research plans, (b) identifying and reducing questionable research practices and researcher biases, (c) distinguishing between confirmatory and exploratory analyses, and, in the case of registered reports, (d) facilitating results-blind peer review, and (e) reducing publication bias.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322747+00:00" } }, { "group": "Reliability", "title": "Registered report", "problem": "Pre-result review reduces selective publication yet still routes recognition through a journal.", "repair": "Separate prospective design certification from compulsory publication of all completed outcomes.", "probe": "Audit publication and evidence availability across accepted and rejected predictions.", "slug": "registered-report", "number": 40, "source": { "requested_title": "Registered report", "canonical_title": "Registered report", "pageid": 60911430, "revision_id": 1375710612, "revision_timestamp": "2026-09-19T15:00:18Z", "source_url": "https://en.wikipedia.org/wiki/Registered_report", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375710612", "intro": "A registered report is a scholarly publishing format in which the research question, study design, and analysis plan are peer-reviewed before data are collected or analyzed. This publication model aims to reduce publication bias and improve research transparency by restricting the evaluation of scientific manuscripts to the research question, method, and analysis plan, without consideration of whether the results support the research question or not.\nRegistered reports are a form of preregistration.", "verified": true, "retrieved_at": "2026-10-14T20:47:03.322751+00:00" } }, { "group": "Reliability", "title": "Systematic review", "problem": "Reviews depend on slow search, manual screening and publication access.", "repair": "Maintain continuously refreshed evidence graphs with traceable inclusion decisions and human-auditable exclusions.", "probe": "Compare living automated review coverage with independently curated gold standards.", "slug": "systematic-review", "number": 41, "source": { "requested_title": "Systematic review", "canonical_title": "Systematic review", "pageid": 2994579, "revision_id": 1372813059, "revision_timestamp": "2026-09-02T11:39:33Z", "source_url": "https://en.wikipedia.org/wiki/Systematic_review", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1372813059", "intro": "A systematic review is a scholarly synthesis of the evidence on a clearly presented topic using critical methods to identify, define and assess research on the topic. A systematic review extracts and interprets data from published studies on the topic (in the scientific literature), then analyzes, describes, critically appraises and summarizes interpretations into a refined evidence-based conclusion. For example, a systematic review of randomized controlled trials is a way of summarizing and implementing evidence-based medicine. Systematic reviews, sometimes along with meta-analyses, are generally considered the highest level in the hierarchy of evidence in scientific and medical research.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211301+00:00" } }, { "group": "Reliability", "title": "Research integrity", "problem": "Rules around misconduct do not by themselves guarantee an inspectable chain of custody.", "repair": "Sign data acquisition events, separate execution from verification and alert on evidence discontinuities.", "probe": "Run provenance tampering exercises while preserving due process for contested alerts.", "slug": "research-integrity", "number": 42, "source": { "requested_title": "Research integrity", "canonical_title": "Scientific integrity", "pageid": 36238592, "revision_id": 1377152310, "revision_timestamp": "2026-09-28T02:52:35Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_integrity", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1377152310", "intro": "Research integrity or scientific integrity is an aspect of research ethics that deals with best practice or rules of professional practice of scientists.\nFirst introduced in the 19th century by Charles Babbage, the concept of research integrity came to the fore in the late 1970s. A series of publicized scandals in the United States led to heightened debate on the ethical norms of sciences and the limitations of the self-regulation processes implemented by scientific communities and institutions. Formalized definitions of scientific misconduct, and codes of conduct, became the main policy response after 1990.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211316+00:00" } }, { "group": "Reliability", "title": "Research transparency", "problem": "Transparency is treated as an author practice rather than a property enforced throughout execution.", "repair": "Require stage-level machine-verifiable provenance, not merely supplementary files.", "probe": "Count how many conclusions survive independent replay with no private clarifications.", "slug": "research-transparency", "number": 43, "source": { "requested_title": "Research transparency", "canonical_title": "Research transparency", "pageid": 70984019, "revision_id": 1375658312, "revision_timestamp": "2026-09-19T04:42:27Z", "source_url": "https://en.wikipedia.org/wiki/Research_transparency", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375658312", "intro": "Research transparency is a major aspect of scientific research. It covers a variety of scientific principles and practices: reproducibility, data and code sharing, citation standards or verifiability.\nThe definitions and norms of research transparency significantly differ depending on the disciplines and fields of research. Due to the lack of consistent terminology, research transparency has frequently been defined negatively by addressing non-transparent usages (which are part of questionable research practices).", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211320+00:00" } }, { "group": "Experimentation", "title": "Design of experiments", "problem": "Fixed-factor plans are often selected before observing the full information structure of a problem.", "repair": "Choose experiments adaptively to maximize safe expected discrimination among competing models.", "probe": "Compare adaptive and factorial designs under the same measurement budget.", "slug": "design-of-experiments", "number": 44, "source": { "requested_title": "Design of experiments", "canonical_title": "Design of experiments", "pageid": 9541, "revision_id": 1379887652, "revision_timestamp": "2026-10-12T13:21:55Z", "source_url": "https://en.wikipedia.org/wiki/Design_of_experiments", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379887652", "intro": "The design of experiments (DOE), also known as experimental design, refers to the construction of procedures that attempt to explain how changes in one aspect of a system will lead to changes in other aspects of a system. In general, the design of experiments involves decisions about which aspects of the system to change and which to control based on hypotheses about the sources of variance in the aspects of the system considered by the experimenter. \nDOE is generally associated with experiments where the design introduces conditions that directly affect the variation, but DOE may also refer to the design of quasi-experiments, in which natural conditions that influence the variation are selected for observation.\nIn its simplest form, an experiment aims at predicting the outcome by introducing a change of the preconditions, which is represented by one or more independent variables, also referred to as \"input variables\" or \"predictor variables.\" The change in one or more independent variables is generally hypothesized to result in a change in one or more dependent variables, also referred to as \"output variables\" or \"response variables.\" The experimental design may also identify control variables that must be held constant to prevent external factors from affecting the results.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211324+00:00" } }, { "group": "Experimentation", "title": "Randomized controlled trial", "problem": "Randomization helps identification but can be undermined by noncompliance, attrition and poor transport.", "repair": "Represent assignment, adherence, outcomes and target-population transfer in explicit causal graphs.", "probe": "Audit intent-to-treat and per-protocol claims across realistic missingness scenarios.", "slug": "randomized-controlled-trial", "number": 45, "source": { "requested_title": "Randomized controlled trial", "canonical_title": "Randomized controlled trial", "pageid": 163180, "revision_id": 1350325633, "revision_timestamp": "2026-04-21T07:55:36Z", "source_url": "https://en.wikipedia.org/wiki/Randomized_controlled_trial", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1350325633", "intro": "A randomized controlled trial (RCT) is a type of statistical experiment designed to evaluate the efficacy or safety of an intervention by minimizing bias through the random allocation of participants to one or more comparison groups. \nIn this approach, at least one group receives the intervention or process under study (such as a drug, surgical procedure, medical device or diet), while the other groups receive an alternative treatment, a placebo, or standard care. \nRCTs are a fundamental methodology in modern clinical trials and have been widely considered one of the highest-quality sources of evidence in evidence-based medicine, due to their ability to reduce selection bias and the influence of confounding factors. However, they have also been criticized for failing to reduce bias in some cases.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211327+00:00" } }, { "group": "Experimentation", "title": "Scientific control", "problem": "A control is specified locally but often lacks explicit causal justification.", "repair": "Generate control designs from the causal claims being tested and challenge hidden contamination paths.", "probe": "Test proposed controls against known interference and confounding structures.", "slug": "scientific-control", "number": 46, "source": { "requested_title": "Scientific control", "canonical_title": "Scientific control", "pageid": 1640288, "revision_id": 1375480330, "revision_timestamp": "2026-09-18T01:24:28Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_control", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375480330", "intro": "A scientific control is an element of an experiment or observation designed to minimize the influence of variables other than the independent variable under investigation, thereby reducing the risk of confounding. \nThe use of controls increases the reliability and validity of results by providing a baseline for comparison between experimental measurements and control measurements. In many designs, the control group does not receive the experimental treatment, allowing researchers to isolate the effect of the independent variable. \nScientific controls are a fundamental part of the scientific method, particularly in fields such as biology, chemistry, medicine, and psychology, where complex systems are subject to multiple interacting variables.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211331+00:00" } }, { "group": "Experimentation", "title": "Blinded experiment", "problem": "Blinding is fragile, and unblinding is frequently insufficiently measured.", "repair": "Enforce access separations, check leaks and log unblinding events without compromising participant safety.", "probe": "Inject subtle treatment leaks and measure whether systems identify them.", "slug": "blinded-experiment", "number": 47, "source": { "requested_title": "Blinded experiment", "canonical_title": "Blinded experiment", "pageid": 277248, "revision_id": 1357239547, "revision_timestamp": "2026-06-01T14:58:40Z", "source_url": "https://en.wikipedia.org/wiki/Blinded_experiment", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1357239547", "intro": "In a blind or blinded experiment, information that could influence participants or investigators is withheld until the experiment is completed. Blinding is used to reduce or eliminate potential sources of bias, such as participants’ expectations, the observer-expectancy effect, observer bias, confirmation bias, and other cognitive or procedural influences. \nBlinding can be applied to different participants in an experiment, including study subjects, researchers, technicians, data analysts, and outcome assessors. When multiple groups are blinded simultaneously (for example, both participants and researchers), the design is referred to as a double-blind study.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211334+00:00" } }, { "group": "Experimentation", "title": "Natural experiment", "problem": "Apparently exogenous exposure may hide violations of the identification assumptions.", "repair": "Continuously audit exclusion, timing, manipulation and parallel-trend assumptions.", "probe": "Stress-test natural-experiment claims with adversarial synthetic counterexamples.", "slug": "natural-experiment", "number": 48, "source": { "requested_title": "Natural experiment", "canonical_title": "Natural experiment", "pageid": 170141, "revision_id": 1348959741, "revision_timestamp": "2026-04-15T03:28:28Z", "source_url": "https://en.wikipedia.org/wiki/Natural_experiment", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1348959741", "intro": "A natural experiment is a study in which individuals are exposed to experimental and control conditions that are determined by nature or by other factors outside the control of the investigators. The exposure process may resemble random assignment. Thus, natural experiments are observational studies and are not controlled in the traditional sense of a randomized experiment (an intervention study).\nNatural experiments are generally more reliable when there is a clearly defined exposure or intervention that affects a well-defined subpopulation (with a comparable subpopulation remaining unexposed), such that differences in outcomes may be attributed to the exposure or intervention.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211338+00:00" } }, { "group": "Experimentation", "title": "Observational study", "problem": "Large observational datasets invite confident causal narratives without identifying interventions.", "repair": "Require an explicit estimand, causal graph and sensitivity analysis before asserting effects.", "probe": "Compare inferred intervention effects with hidden randomized ground truth.", "slug": "observational-study", "number": 49, "source": { "requested_title": "Observational study", "canonical_title": "Observational study", "pageid": 6326483, "revision_id": 1346158126, "revision_timestamp": "2026-03-30T07:29:42Z", "source_url": "https://en.wikipedia.org/wiki/Observational_study", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1346158126", "intro": "In fields such as epidemiology, social sciences, psychology and statistics, an observational study draws conclusions without controlling the independent variable due to ethical or practical limitations. One common example studies the effect of a treatment, where the researcher does not assign subjects to treatment or control group. This is in contrast with experiments, such as randomized controlled trials, where each subject is randomly assigned to a treated group or a control group. Observational studies, for lacking an assignment mechanism, naturally present difficulties for inferential analysis.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211341+00:00" } }, { "group": "Experimentation", "title": "Causal inference", "problem": "Models may estimate associations accurately while causal assumptions remain unverifiable.", "repair": "Store causal identification claims separately from estimates and design targeted interventions.", "probe": "Measure how often agents distinguish identification from prediction.", "slug": "causal-inference", "number": 50, "source": { "requested_title": "Causal inference", "canonical_title": "Causal inference", "pageid": 37103476, "revision_id": 1368546517, "revision_timestamp": "2026-08-09T18:31:26Z", "source_url": "https://en.wikipedia.org/wiki/Causal_inference", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1368546517", "intro": "Causal inference is the process of determining the independent, actual effect of a particular phenomenon that is a component of a larger system. The main difference between causal inference and inference of association is that causal inference analyzes the response of an effect variable when a cause of the effect variable is changed.\nThe study of why things occur is called etiology, and can be described using the language of scientific causal notation. Causal inference is said to provide the evidence of causality theorized by causal reasoning.", "verified": true, "retrieved_at": "2026-10-14T20:47:04.211344+00:00" } }, { "group": "Experimentation", "title": "Confounding", "problem": "Unobserved common causes can persist despite sophisticated adjustment.", "repair": "Search competing causal graphs and report assumption-specific identification limits.", "probe": "Inject latent confounders and test the honesty of uncertainty statements.", "slug": "confounding", "number": 51, "source": { "requested_title": "Confounding", "canonical_title": "Confounding", "pageid": 3105999, "revision_id": 1371518385, "revision_timestamp": "2026-08-26T23:12:07Z", "source_url": "https://en.wikipedia.org/wiki/Confounding", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1371518385", "intro": "In causal inference, confounding is a form of systematic error (or bias) that can distort estimates of causal effects in observational studies. A confounder is traditionally understood to be a variable that (1) independently predicts the outcome (or dependent variable), (2) is associated with the exposure (or independent variable), and (3) is not on the causal pathway between the exposure and the outcome. Failure to control for a confounder results in a spurious association between exposure and outcome.\nConfounding is a causal concept rather than a purely statistical one, and therefore cannot be fully described by correlations or associations alone.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840268+00:00" } }, { "group": "Experimentation", "title": "Counterfactual conditional", "problem": "Counterfactual claims require assumptions that cannot be observed directly for each unit.", "repair": "Keep structural counterfactual models separate from observed potential outcomes and expose identifying assumptions.", "probe": "Challenge interventions under several compatible structural models.", "slug": "counterfactual-conditional", "number": 52, "source": { "requested_title": "Counterfactual conditional", "canonical_title": "Counterfactual conditional", "pageid": 491578, "revision_id": 1368679740, "revision_timestamp": "2026-08-10T13:23:23Z", "source_url": "https://en.wikipedia.org/wiki/Counterfactual_conditional", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1368679740", "intro": "Counterfactual conditionals (also contrafactual/subjunctive/X-marked conditionals) are conditional sentences that describe what would have been true if circumstances had been different, typically when the antecedent is taken to be false or incompatible with what actually happened. In English they are often formed with a past or irrealis morphology, as in \"If Peter believed in ghosts, he would be afraid to be here\", and are standardly contrasted with indicative conditionals, which are generally used to discuss live or open possibilities.\nThe name subjunctive conditionals is sometimes preferred because counterfactuals are not always \"contrary to fact\" as the word \"counterfactual\" implies, but the name counterfactual conditionals is sometimes preferred because counterfactuals are not always grammatically subjunctive. Hence the name X-marking has been developed as a compromise, although it is newer in the literature; see § Terminology.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840286+00:00" } }, { "group": "Experimentation", "title": "External validity", "problem": "An internally valid result is routinely projected beyond the population and setting studied.", "repair": "Compute transport conditions and document domain limits as part of every scientific claim.", "probe": "Test projected effects in systematically shifted target populations.", "slug": "external-validity", "number": 53, "source": { "requested_title": "External validity", "canonical_title": "External validity", "pageid": 1900609, "revision_id": 1370590349, "revision_timestamp": "2026-08-22T00:35:31Z", "source_url": "https://en.wikipedia.org/wiki/External_validity", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1370590349", "intro": "External validity is the validity of applying the conclusions of a scientific study outside the context of that study. In other words, it is the extent to which the results of a study can generalize or transport to other situations, people, stimuli, and times. Generalizability refers to the applicability of a predefined sample to a broader population while transportability refers to the applicability of one sample to another target population. In contrast, internal validity is the validity of conclusions drawn within the context of a particular study.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840291+00:00" } }, { "group": "Experimentation", "title": "Internal validity", "problem": "Methodological checklists can miss unexpected pathways from assignment to outcomes.", "repair": "Execute adversarial causal audits over randomization, attrition, measurement and contamination.", "probe": "Benchmark detection against investigator-designed validity threats.", "slug": "internal-validity", "number": 54, "source": { "requested_title": "Internal validity", "canonical_title": "Internal validity", "pageid": 1562189, "revision_id": 1374310121, "revision_timestamp": "2026-09-11T05:30:02Z", "source_url": "https://en.wikipedia.org/wiki/Internal_validity", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1374310121", "intro": "Internal validity is the extent to which a piece of evidence supports a claim about cause and effect, within the context of a particular study. It is one of the most important properties of scientific studies and is an important concept in reasoning about evidence more generally. Internal validity is determined by how well a study can rule out alternative explanations for its findings (usually, sources of systematic error or 'bias'). It contrasts with external validity, the extent to which results can justify conclusions about other contexts (that is, the extent to which results can be generalized).", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840295+00:00" } }, { "group": "Experimentation", "title": "Measurement uncertainty", "problem": "Final error bars frequently omit uncertain calibration and preprocessing decisions.", "repair": "Propagate sensor, preprocessing and model uncertainties along the complete evidence chain.", "probe": "Replay a measurement after known calibration changes and validate updated downstream intervals.", "slug": "measurement-uncertainty", "number": 55, "source": { "requested_title": "Measurement uncertainty", "canonical_title": "Measurement uncertainty", "pageid": 3069520, "revision_id": 1367370535, "revision_timestamp": "2026-08-02T18:11:10Z", "source_url": "https://en.wikipedia.org/wiki/Measurement_uncertainty", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1367370535", "intro": "In metrology, measurement uncertainty is the expression of the statistical dispersion of the values attributed to a quantity measured on an interval or ratio scale. \nAll measurements are subject to uncertainty and a measurement result is complete only when it is accompanied by a statement of the associated uncertainty, such as the standard deviation. By international agreement, this uncertainty has a probabilistic basis and reflects incomplete knowledge of the quantity value. It is a non-negative parameter.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840299+00:00" } }, { "group": "Experimentation", "title": "Adaptive clinical trial", "problem": "Adaptation can boost efficiency but introduces operational bias and statistical complexity.", "repair": "Treat every adaptation as a pre-specified, audited policy with validated error guarantees and ethical oversight.", "probe": "Simulate many treatment-response regimes to confirm operating characteristics.", "slug": "adaptive-clinical-trial", "number": 56, "source": { "requested_title": "Adaptive clinical trial", "canonical_title": "Adaptive design (medicine)", "pageid": 65297106, "revision_id": 1364723345, "revision_timestamp": "2026-07-18T04:01:57Z", "source_url": "https://en.wikipedia.org/wiki/Adaptive_design_(medicine)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1364723345", "intro": "In an adaptive design of a clinical trial, the parameters and conduct of the trial for a candidate drug or vaccine may be changed based on an interim analysis. Adaptive design typically involves advanced statistics to interpret a clinical trial endpoint. This is in contrast to traditional single-arm (i.e. non-randomized) clinical trials or randomized clinical trials (RCTs) that are static in their protocol and do not modify any parameters until the trial is completed.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840303+00:00" } }, { "group": "Verification", "title": "Automated theorem proving", "problem": "Search for proofs is conflated with independently establishing that a theorem statement was proved.", "repair": "Separate conjecture generation, trusted kernel checking and semantic alignment with natural-language claims.", "probe": "Feed altered formal statements to independent semantic auditors and measure misalignment detection.", "slug": "automated-theorem-proving", "number": 57, "source": { "requested_title": "Automated theorem proving", "canonical_title": "Automated theorem proving", "pageid": 2546, "revision_id": 1380246457, "revision_timestamp": "2026-10-14T17:19:07Z", "source_url": "https://en.wikipedia.org/wiki/Automated_theorem_proving", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1380246457", "intro": "Automated theorem proving (ATP), also known as automated deduction, is a subfield of automated reasoning and mathematical logic dealing with proving mathematical theorems by computer programs. Automated reasoning over mathematical proof was a major motivating factor for the development of computer science.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840307+00:00" } }, { "group": "Verification", "title": "Proof assistant", "problem": "Machine-checked derivations remain conditional on encodings, axioms and trusted components.", "repair": "Record kernel versions, trusted assumptions, statement mappings and independent checking receipts.", "probe": "Rebuild proofs under a clean kernel and inspect assumptions and theorem translation.", "slug": "proof-assistant", "number": 58, "source": { "requested_title": "Proof assistant", "canonical_title": "Proof assistant", "pageid": 1258607, "revision_id": 1380223205, "revision_timestamp": "2026-10-14T14:25:13Z", "source_url": "https://en.wikipedia.org/wiki/Proof_assistant", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1380223205", "intro": "In computer science and mathematical logic, a proof assistant or interactive theorem prover is a software tool to assist with the development of formal proofs by human–machine collaboration. This involves some sort of interactive proof editor, or other interface, with which a human can guide the search for proofs, the details of which are stored in, and some steps provided by, a computer.\nA recent effort within this field is making these tools use artificial intelligence to automate the formalization of ordinary mathematics.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840310+00:00" } }, { "group": "Verification", "title": "Formal verification", "problem": "Proving software properties may ignore a mistaken specification or an external physical failure.", "repair": "Verify specifications against use cases and maintain distinct model, implementation and operating-environment claims.", "probe": "Deliberately weaken specifications and test whether external validation catches them.", "slug": "formal-verification", "number": 59, "source": { "requested_title": "Formal verification", "canonical_title": "Formal verification", "pageid": 270054, "revision_id": 1377238699, "revision_timestamp": "2026-09-28T12:58:59Z", "source_url": "https://en.wikipedia.org/wiki/Formal_verification", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1377238699", "intro": "In the context of hardware and software systems, formal verification is the act of proving or disproving the correctness of a system with respect to a certain formal specification or property, using formal methods of mathematics. \nFormal verification is a key incentive for formal specification of systems, and is at the core of formal methods.\nIt represents an important dimension of analysis and verification in electronic design automation and is one approach to software verification. The use of formal verification enables the highest Evaluation Assurance Level (EAL7) in the framework of common criteria for computer security certification.", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840314+00:00" } }, { "group": "Verification", "title": "Lean (proof assistant)", "problem": "The presence of Lean code can be mistaken for an independent certification of the intended mathematics.", "repair": "Pin exact statements, packages, kernel trust and informal-to-formal correspondence checks.", "probe": "Audit theorem statements for changed quantifiers and hidden added hypotheses.", "slug": "lean-proof-assistant", "number": 60, "source": { "requested_title": "Lean (proof assistant)", "canonical_title": "Lean (proof assistant)", "pageid": 62889984, "revision_id": 1374563227, "revision_timestamp": "2026-09-12T20:28:36Z", "source_url": "https://en.wikipedia.org/wiki/Lean_(proof_assistant)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1374563227", "intro": "Lean is a proof assistant and a functional programming language. It is based on the calculus of constructions with inductive types (specifically, the Calculus of Inductive Constructions), the foundational type theory developed with the Coq theorem prover, which was renamed to Rocq in 2024. It is a free and open-source software project hosted on GitHub. Development is currently supported by the nonprofit Lean Focused Research Organization (FRO).", "verified": true, "retrieved_at": "2026-10-14T20:47:05.840317+00:00" } }, { "group": "Verification", "title": "Computational science", "problem": "Numerical results can inherit uncontrolled approximation, environment and implementation choices.", "repair": "Make solvers, tolerances, seeds, numerical stability and verification experiments replayable.", "probe": "Cross-check results using independent numerical methods and precision budgets.", "slug": "computational-science", "number": 61, "source": { "requested_title": "Computational science", "canonical_title": "Computational science", "pageid": 1181008, "revision_id": 1376349250, "revision_timestamp": "2026-09-23T16:30:09Z", "source_url": "https://en.wikipedia.org/wiki/Computational_science", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376349250", "intro": "Computational science, also known as scientific computing, technical computing or scientific computation (SC), is a division of science, and more specifically the computer sciences, which uses advanced computing capabilities to understand and solve complex physical problems in science. While this typically extends into computational specializations, this field of study includes: \n\nAlgorithms (numerical and non-numerical): mathematical models, computational models, and computer simulations developed to solve sciences (e.g, physical, biological, and social), engineering, and humanities problems\nComputer hardware that develops and optimizes the advanced system hardware, firmware, networking, and data management components needed to solve computationally demanding problems\nThe computing infrastructure that supports both the science and engineering problem solving and the developmental computer and information science\nIn practical use, it is typically the application of computer simulation and other forms of computation from numerical analysis and theoretical computer science to solve problems in various scientific disciplines. The field is different from theory and laboratory experiments, which are the traditional forms of science and engineering. The scientific computing approach is to gain understanding through the analysis of mathematical models implemented on computers.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763205+00:00" } }, { "group": "Verification", "title": "Scientific computing", "problem": "A successful computation is often reported without enough executable dependency detail.", "repair": "Use hermetic build graphs, content-addressed inputs and reproducibility receipts.", "probe": "Recompute published outputs with pinned environments on different hardware.", "slug": "scientific-computing", "number": 62, "source": { "requested_title": "Scientific computing", "canonical_title": "Computational science", "pageid": 1181008, "revision_id": 1376349250, "revision_timestamp": "2026-09-23T16:30:09Z", "source_url": "https://en.wikipedia.org/wiki/Computational_science", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1376349250", "intro": "Computational science, also known as scientific computing, technical computing or scientific computation (SC), is a division of science, and more specifically the computer sciences, which uses advanced computing capabilities to understand and solve complex physical problems in science. While this typically extends into computational specializations, this field of study includes: \n\nAlgorithms (numerical and non-numerical): mathematical models, computational models, and computer simulations developed to solve sciences (e.g, physical, biological, and social), engineering, and humanities problems\nComputer hardware that develops and optimizes the advanced system hardware, firmware, networking, and data management components needed to solve computationally demanding problems\nThe computing infrastructure that supports both the science and engineering problem solving and the developmental computer and information science\nIn practical use, it is typically the application of computer simulation and other forms of computation from numerical analysis and theoretical computer science to solve problems in various scientific disciplines. The field is different from theory and laboratory experiments, which are the traditional forms of science and engineering. The scientific computing approach is to gain understanding through the analysis of mathematical models implemented on computers.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763223+00:00" } }, { "group": "Verification", "title": "Scientific workflow system", "problem": "Workflow orchestration is optimized for throughput rather than epistemic traceability.", "repair": "Use typed evidence transitions that refuse invalid provenance or confirmation-data reuse.", "probe": "Inject missing or mutated workflow steps and verify that claims become invalidated.", "slug": "scientific-workflow-system", "number": 63, "source": { "requested_title": "Scientific workflow system", "canonical_title": "Scientific workflow system", "pageid": 26762608, "revision_id": 1369246739, "revision_timestamp": "2026-08-13T20:00:53Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_workflow_system", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1369246739", "intro": "A scientific workflow system is a specialized form of a workflow management system designed specifically to compose and execute a series of computational or data manipulation steps, or workflow, in a scientific application. Scientific workflow systems are generally developed for use by scientists from different disciplines like astronomy, earth science, and bioinformatics. All such systems are based on an abstract representation of how a computation proceeds in the form of a directed graph, where each node represents a task to be executed and edges represent either data flow or execution dependencies between different tasks. Each system typically provides a visual front-end, allowing the user to build and modify complex applications with little or no programming expertise.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763229+00:00" } }, { "group": "Verification", "title": "Laboratory automation", "problem": "Automated experimentation still risks optimizing a flawed objective or unreliable instruments.", "repair": "Pair robotic experimenters with independent adversarial design and calibration agents.", "probe": "Compare autonomous and human-directed programs under equivalent safe lab budgets.", "slug": "laboratory-automation", "number": 64, "source": { "requested_title": "Laboratory automation", "canonical_title": "Laboratory automation", "pageid": 5670694, "revision_id": 1377714125, "revision_timestamp": "2026-09-30T19:10:28Z", "source_url": "https://en.wikipedia.org/wiki/Laboratory_automation", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1377714125", "intro": "Laboratory automation is the use of automated equipment, robotics and software to perform or coordinate processes in a laboratory with reduced human intervention. Automation can range from individual instruments that perform specific tasks to integrated systems that coordinate multiple stages of an experimental or analytical workflow. It is used in areas including clinical testing, high-throughput screening, biotechnology, chemistry and materials science.\nLaboratory automation can increase experimental throughput, reduce repetitive manual work, and improve the consistency of laboratory processes.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763235+00:00" } }, { "group": "Verification", "title": "Active learning (machine learning)", "problem": "Selecting useful labels can exploit dataset artifacts while neglecting scientific identifiability.", "repair": "Choose samples and interventions for causal discrimination and domain coverage rather than score alone.", "probe": "Benchmark active learners when correlations invert in unseen regimes.", "slug": "active-learning-machine-learning", "number": 65, "source": { "requested_title": "Active learning (machine learning)", "canonical_title": "Active learning (machine learning)", "pageid": 28801798, "revision_id": 1378688515, "revision_timestamp": "2026-10-05T18:07:22Z", "source_url": "https://en.wikipedia.org/wiki/Active_learning_(machine_learning)", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1378688515", "intro": "Active learning is a special case of machine learning in which a learning algorithm can interactively query a human user (or some other information source) to label new data points with the desired outputs. The human user must possess expertise in the problem domain, including the ability to consult authoritative sources when necessary. In statistics literature, it is sometimes also called optimal experimental design. The information source is also called teacher or oracle.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763241+00:00" } }, { "group": "Institutions", "title": "Research ethics", "problem": "Ethical evaluation can appear as a one-time approval external to the research execution.", "repair": "Keep consent, risk thresholds, participant protections and human authorization executable and auditable.", "probe": "Test that every prohibited experimental action is blocked while authorized work remains possible.", "slug": "research-ethics", "number": 66, "source": { "requested_title": "Research ethics", "canonical_title": "Research ethics", "pageid": 3609425, "revision_id": 1368094508, "revision_timestamp": "2026-08-07T00:35:39Z", "source_url": "https://en.wikipedia.org/wiki/Research_ethics", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1368094508", "intro": "Research ethics is a discipline within the study of applied ethics. Its scope ranges from general scientific integrity and misconduct to the treatment of human and animal subjects. The social responsibilities of scientists and researchers are not traditionally included and are less well defined.\nThe discipline is most developed in medical research.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763246+00:00" } }, { "group": "Institutions", "title": "Scientific journal", "problem": "Editorial acceptance becomes a social gateway through which scientific results obtain visibility.", "repair": "Separate dissemination and explanatory curation from the machine-readable record of claims and evidence.", "probe": "Measure the time from valid new evidence to its availability without journal mediation.", "slug": "scientific-journal", "number": 67, "source": { "requested_title": "Scientific journal", "canonical_title": "Academic journal", "pageid": 324572, "revision_id": 1379833314, "revision_timestamp": "2026-10-12T04:39:57Z", "source_url": "https://en.wikipedia.org/wiki/Academic_journal", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379833314", "intro": "An academic journal or scholarly journal is a periodical publication in which scholarship relating to a particular academic discipline is published. They serve as permanent and transparent forums for the dissemination, scrutiny, and discussion of research. Unlike professional magazines or trade magazines, the articles are mostly written by researchers rather than staff writers employed by the journal. They nearly universally require peer review for research articles or other scrutiny from contemporaries competent and established in their respective fields.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763250+00:00" } }, { "group": "Institutions", "title": "Academic publishing", "problem": "Article incentives often reward narratively tidy novelty rather than cumulative correction.", "repair": "Publish versioned evidence objects, including negative outcomes, dependency updates and competing interpretations.", "probe": "Compare error-correction latency in paper-first and evidence-first workflows.", "slug": "academic-publishing", "number": 68, "source": { "requested_title": "Academic publishing", "canonical_title": "Academic publishing", "pageid": 324570, "revision_id": 1375045430, "revision_timestamp": "2026-09-15T14:54:43Z", "source_url": "https://en.wikipedia.org/wiki/Academic_publishing", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375045430", "intro": "Academic publishing is the subfield of publishing which distributes academic research and scholarship. Most academic work is published in academic journal articles, books or theses. The part of academic written output that is not formally published but merely printed up or posted on the Internet is often called \"grey literature\". Most scientific and scholarly journals, and many academic and scholarly books, though not all, are based on some form of peer review or editorial refereeing to qualify texts for publication.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763253+00:00" } }, { "group": "Institutions", "title": "Bibliometrics", "problem": "Citation volume measures attention, not independent correctness or the value of error detection.", "repair": "Develop multidimensional evidence-quality metrics without collapsing them into a single prestige score.", "probe": "Audit whether metrics reward detecting flaws more than generating prolific claims.", "slug": "bibliometrics", "number": 69, "source": { "requested_title": "Bibliometrics", "canonical_title": "Bibliometrics", "pageid": 1223245, "revision_id": 1377099702, "revision_timestamp": "2026-09-27T22:03:09Z", "source_url": "https://en.wikipedia.org/wiki/Bibliometrics", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1377099702", "intro": "Bibliometrics is the application of statistical methods to the study of bibliographic data, especially in scientific and library and information science contexts. It is closely associated with scientometrics (the analysis of scientific metrics and indicators), to the point that both fields largely overlap.\nBibliometrics studies first appeared in the late 19th century. They have known a significant development after the Second World War in a context of \"periodical crisis\" and new technical opportunities offered by computing tools.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763257+00:00" } }, { "group": "Institutions", "title": "Citation analysis", "problem": "Citation networks may propagate untested claims and self-referential chains of authority.", "repair": "Annotate citation edges by evidentiary function, independence and actual claim support.", "probe": "Detect when downstream papers cite a source for a proposition it never established.", "slug": "citation-analysis", "number": 70, "source": { "requested_title": "Citation analysis", "canonical_title": "Citation analysis", "pageid": 1235972, "revision_id": 1370697715, "revision_timestamp": "2026-08-22T16:10:45Z", "source_url": "https://en.wikipedia.org/wiki/Citation_analysis", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1370697715", "intro": "Citation analysis is the examination of the frequency, patterns, and graphs of citations in documents. It uses the directed graph of citations – links from one document to another document – to reveal properties of the documents. A typical aim would be to identify the most important documents in a collection. A classic example is that of the citations between academic articles and books.", "verified": true, "retrieved_at": "2026-10-14T20:47:06.763260+00:00" } }, { "group": "Institutions", "title": "Open data", "problem": "Making data downloadable does not guarantee lawful reuse, sufficient context or interpretability.", "repair": "Attach machine-readable permissions, consent boundaries, measurement provenance and validation tests.", "probe": "Evaluate independent reconstruction of variables and meaningful error bounds.", "slug": "open-data", "number": 71, "source": { "requested_title": "Open data", "canonical_title": "Open data", "pageid": 7697770, "revision_id": 1380195714, "revision_timestamp": "2026-10-14T09:52:59Z", "source_url": "https://en.wikipedia.org/wiki/Open_data", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1380195714", "intro": "Open data are data that are openly accessible, exploitable, editable and shareable by anyone for any purpose. Open data are generally licensed under an open license.\nThe goals of the open data movement are similar to those of other \"open(-source)\" movements such as open-source software, open-source hardware, open content, open specifications, open education, open educational resources, open government, open knowledge, open access, open science, and the open web. The growth of the open data movement is paralleled by a rise in intellectual property rights.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705135+00:00" } }, { "group": "Institutions", "title": "FAIR data", "problem": "Findability and interoperability principles can be satisfied superficially without epistemic reproducibility.", "repair": "Connect FAIR metadata to executable derivations, caveats and independent evidence checks.", "probe": "Run cross-repository retrieval and end-to-end derivation acceptance tests.", "slug": "fair-data", "number": 72, "source": { "requested_title": "FAIR data", "canonical_title": "FAIR data", "pageid": 58255600, "revision_id": 1370308559, "revision_timestamp": "2026-08-20T10:23:24Z", "source_url": "https://en.wikipedia.org/wiki/FAIR_data", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1370308559", "intro": "FAIR data is data which meets the 2016 FAIR principles of findability, accessibility, interoperability, and reusability, first formally published in 2016 .\nThe FAIR principles emphasise machine-actionability — the capacity of computational systems to find, access, interoperate, and reuse data with none or minimal human intervention — because humans increasingly rely on computational support to deal with data as a result of the increase in the volume, complexity, and rate of production of data. The principles were designed to apply not only to data in the conventional sense, but also to the algorithms, tools, and workflows that led to that data: all scholarly digital research objects benefit from application of these principles, since all components of the research process must be available to ensure transparency, reproducibility, and reusability.\nThe abbreviation FAIR/O data is sometimes used to indicate that the dataset or database in question complies with the FAIR principles and also carries an explicit data‑capable open license.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705155+00:00" } }, { "group": "Institutions", "title": "Citizen science", "problem": "Mass participation can increase coverage yet introduce unequal measurement quality and overlooked local expertise.", "repair": "Model contributor-specific uncertainty while preserving participant agency, credit and contestability.", "probe": "Compare calibrated mixed-observer evidence with naive aggregation.", "slug": "citizen-science", "number": 73, "source": { "requested_title": "Citizen science", "canonical_title": "Citizen science", "pageid": 2155752, "revision_id": 1375457912, "revision_timestamp": "2026-09-17T22:19:53Z", "source_url": "https://en.wikipedia.org/wiki/Citizen_science", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1375457912", "intro": "Citizen science is research conducted with the participation of the general public, amateur or nonprofessional researchers, or participants from the fields of science, social science, and many other disciplines. The exact definition of \"citizen science\" varies, with different individuals and organizations having their own specific interpretations of its scope. Citizen science is employed in a wide range of areas of study, including ecology, biology, conservation, health and medical research, astronomy, media and communications, and information science.\nThe applications and functions of citizen science in research projects are multifaceted.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705160+00:00" } }, { "group": "Institutions", "title": "Scientific community", "problem": "Institutional and disciplinary silos impede recombination of findings into common models.", "repair": "Replace document handoffs with shared claim identifiers, federated provenance and contestable forks.", "probe": "Measure the fraction of cross-field dependencies resolved without manual transcription.", "slug": "scientific-community", "number": 74, "source": { "requested_title": "Scientific community", "canonical_title": "Scientific community", "pageid": 338705, "revision_id": 1369270996, "revision_timestamp": "2026-08-13T23:44:51Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_community", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1369270996", "intro": "The scientific community is a diverse network of interacting scientists. It includes many \"sub-communities\" working on particular scientific fields, and within particular institutions; interdisciplinary and cross-institutional activities are also significant. Objectivity is expected to be achieved by the scientific method. Peer review, through discussion and debate within journals and conferences, assists in this objectivity by maintaining the quality of research methodology and interpretation of results.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705175+00:00" } }, { "group": "Institutions", "title": "Research misconduct", "problem": "Fabrication and selective disclosure can survive polished papers and plausible machine-generated narratives.", "repair": "Require signed observation chains and independent replication while maintaining fair investigation procedures.", "probe": "Run synthetic tampering challenges to quantify false positives and missed manipulation.", "slug": "research-misconduct", "number": 75, "source": { "requested_title": "Research misconduct", "canonical_title": "Scientific misconduct", "pageid": 29537, "revision_id": 1379337313, "revision_timestamp": "2026-10-09T09:40:41Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_misconduct", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379337313", "intro": "Scientific misconduct is the violation of ethical and professional standards in research, including fabrication, falsification, plagiarism, and other practices that compromise the integrity of the design, conduct, analysis, reporting, or publication of scientific or research findings.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705217+00:00" } }, { "group": "Institutions", "title": "Science studies", "problem": "Institutions, power and laboratory practice influence what counts as knowledge, but their effects are rarely modeled operationally.", "repair": "Represent funding, gatekeeping, incentives and social selection as empirical parts of the scientific pipeline.", "probe": "Evaluate whether simulated incentive shifts predict observed research distortions.", "slug": "science-studies", "number": 76, "source": { "requested_title": "Science studies", "canonical_title": "Science studies", "pageid": 358601, "revision_id": 1373996015, "revision_timestamp": "2026-09-09T04:47:23Z", "source_url": "https://en.wikipedia.org/wiki/Science_studies", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1373996015", "intro": "Science studies is an interdisciplinary research area that seeks to situate scientific expertise in broad social, historical, and philosophical contexts. It uses various methods to analyze the production, representation and reception of scientific knowledge and its epistemic and semiotic role.\nSimilarly to cultural studies, science studies are defined by the subject of their research and encompass a large range of different theoretical and methodological perspectives and practices. The interdisciplinary approach may include and borrow methods from the humanities, natural and formal sciences, from scientometrics to ethnomethodology or cognitive science.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705235+00:00" } }, { "group": "Institutions", "title": "Scientific consensus", "problem": "Agreement among experts is useful but cannot replace the underlying evidence and conditional inference.", "repair": "Expose why models agree, where they diverge, and which experimental results support each confidence claim.", "probe": "Separate consensus formation tests from empirical reliability benchmarks.", "slug": "scientific-consensus", "number": 77, "source": { "requested_title": "Scientific consensus", "canonical_title": "Scientific consensus", "pageid": 423727, "revision_id": 1330333272, "revision_timestamp": "2025-12-30T20:57:28Z", "source_url": "https://en.wikipedia.org/wiki/Scientific_consensus", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1330333272", "intro": "Scientific consensus is the collective judgment, position, and opinion of the vast majority of active, qualified experts on a conclusion in a specific scientific discipline. Scientific consensus results from the self-correcting scientific process of peer review, replication of the event through the scientific method, scholarly debate, meta-analysis, and publication of high-quality review articles, monographs, or guidelines in reputable books and journals to establish facts and durable knowledge about the topic.\nReaching consensus requires significant scientific agreement among qualified experts, a process based on scientific substantiation of a claim that meets the burden of proof by proposing a possible cause-and-effect mechanism supported by the totality of evidence, leading to agreement among experts. In many countries, scientific consensus established on significant scientific agreement is the basis for regulatory approval of drugs to specify a health claim for the properties of the approved therapeutic agent.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705242+00:00" } }, { "group": "Institutions", "title": "Goodhart's law", "problem": "Optimizing publication counts or benchmark scores can corrupt the underlying purpose of science.", "repair": "Maintain plural quality signals, adversarial audits and periodically replaced evaluation sets.", "probe": "Compare systems that optimize a single score with systems assessed on unseen evidence quality.", "slug": "goodhart-s-law", "number": 78, "source": { "requested_title": "Goodhart's law", "canonical_title": "Goodhart's law", "pageid": 159956, "revision_id": 1379318143, "revision_timestamp": "2026-10-09T05:33:59Z", "source_url": "https://en.wikipedia.org/wiki/Goodhart%27s_law", "permalink": "https://en.wikipedia.org/w/index.php?oldid=1379318143", "intro": "Goodhart's law is an adage that has been stated as, \"When a measure becomes a target, it ceases to be a good measure\". It is named after British economist Charles Goodhart, who is credited with expressing the core idea of the adage in a 1975 article on monetary policy in the United Kingdom:\n\nAny observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.", "verified": true, "retrieved_at": "2026-10-14T20:47:07.705280+00:00" } } ] }