Pith. sign in

REVIEW 4 major objections 6 minor 58 references

TRACES: A Benchmark for Epistemic Reliability in Scientific Reasoning by LLMs

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Across 30 models and 42 flawed papers, language models design follow-up research on false premises in 95% of non-empty responses.

desk verdict TRACES measures a real failure mode—models build study designs on retracted or incoherent premises—but the 93% engagement number needs an independent audit of probe boundaries. read the letter →

arxiv 2608.11415 v1 pith:PCARYHQB submitted 2026-08-11 cs.IR cs.AI

Valentin Rodionov , Shamil Assylbekov This is my paper · ORCID
classification cs.IRcs.AI
keywords epistemicreliabilityscientificreasoninglargelanguagemodelsinfluencefailurerateretractedpaperspseudoscienceagenticsciencebenchmark
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

TRACES tries to establish that large language models, when handed the framing of a retracted, fraudulent, or pseudoscientific paper in a plausible research context, will mostly design follow-up work on top of the false premise instead of refusing. Across 30 models and 10 repeated runs on a 42-probe corpus, only 7% of responses refused outright, giving an aggregate agentic failure rate of 0.93; even after crediting content-specific warnings, the interactive failure rate is 0.809. The paper argues this is not a knowledge defect that more data fixes but a behavioral pattern: refusals concentrate on a handful of famous or flamboyantly pseudoscientific topics and disappear on structurally similar but obscure papers. If the result holds, no current model is safe to deploy as an unsupervised scientific research agent.

What carries the argument

The instrument is the TRACES probe: a near-verbatim preamble from an unreliable paper, a plausible study-design request with no legitimate answer unless the flawed premise is accepted, and withheld paper-specific details scored by level. The central metrics are the two Influence Failure Rates — IFR-a (agentic: refusal is the only pass) and IFR-i (interactive: engagement with a content-specific warning also passes) — plus the Engagement Depth Index (EDI), which measures how much paper- or field-specific withheld vocabulary the model reproduces. The cross-tabulation of REFUSED/RECOGNIZED signals is what lets the paper separate 'safe because it reasoned' from 'safe because a classifier blitzed on a notorious topic,' and the gap between IFR-a and IFR-i quantifies the disclaimer tax.

What would settle it

Ask independent domain experts whether each of the 42 operational requests admits a legitimate non-accepting answer, such as an explicitly hypothetical study design or a critical analysis of the premise; if even a handful do, the engagement rate would no longer measure premise acceptance. A quicker check is to run the same probes against matched control papers with sound premises — the paper reports that refusals disappear under such controls, so if control refusals rose, the interpretation would collapse.

Watch

Extended reading notes

Core claim

The paper's central claim is that scientific reasoning benchmarks that score correct answers cannot see the failure mode that matters for agentic science: a model that fluently adopts and extends an unsound premise. The authors build a probe corpus from 42 unreliable papers, each probe pairing near-verbatim text from the paper with a first-person request to design a follow-up study whose every reasonable path requires accepting the flawed premise. They measure two failure rates: IFR-a, where only outright refusal counts as a pass, and IFR-i, where engagement after a content-specific warning also passes. Across the panel, aggregate IFR-a is 0.93 ± 0.004 and IFR-i is 0.809 ± 0.009, with 95% of non-empty responses engaging with untenable premises; every model fails more than 71% of agentic probes and 22 of 30 fail more than 90%. The authors conclude that the observed refusals are consistent with topic-keyed safety filters rather than epistemic competence, and that credibility assessment must become shared infrastructure for scientific deployment of language models.

Load-bearing premise

The load-bearing premise is that every operational request is genuinely unanswerable unless the model accepts the target paper's false premise, so any non-refusal counts as premise acceptance; if a request could be legitimately answered as a hypothetical, a critique, or a sanity check, the 93% engagement rate overstates the failure.

Editorial extensions

If this is right

  • No model in the 30-model panel refused often enough to be deployed as an unsupervised research agent; even the best interactive models warned users in less than half of responses.
  • A model that produces a 'rigorous double-blind trial of homeopathy' has not produced a safe agentic output, so downstream agents consuming response bodies inherit the false premise regardless of disclaimers.
  • Refusals that do occur cluster on a few high-notoriety topics and largely disappear on structurally similar but obscure papers, implying current safety is topic-keyed rather than epistemic.
  • EDI results indicate contamination is mostly field-level: models across all sizes reproduce field vocabulary, and larger models do so more fluently, so the failure is not fixed by model scale alone.
  • Because 81% of responses contained no warning at all, the paper concludes that credibility annotations and retrieval-based credibility checking are needed infrastructure for scientific LLM deployment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the refusals are topic-keyed, a benchmark that samples famous retractions will overestimate epistemic reliability; the same models would likely engage with unfamiliar but equally unsound papers, so TRACES' mix of notoriety levels is what makes its aggregate meaningful.
  • A natural follow-up the authors leave open is prompt-level mitigation: adding an explicit instruction to assess source reliability could change IFR-a, and measuring whether the improvement is uniform across claim types would separate guardrail effects from genuine reasoning.
  • The single-shot design suggests a multi-turn test: after a model refuses or blanks, re-asking with rephrased framing would reveal whether the refusal is a stable epistemic decision or a shallow filter that can be talked past.
  • A direct deployment test would give a retrieval agent the retraction status of each target paper at inference time; if IFR-a barely moves despite the retrieved evidence, then credibility signals must be enforced by guardrails rather than left to the model's judgment.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper introduces TRACES, a benchmark and methodology for measuring whether LLMs engaged as scientific agents can recognize and refuse to build on unreliable scientific literature. A corpus of 42 probes derived from retracted, fraudulent, and pseudoscientific papers pairs near-verbatim preambles with operational study-design requests; responses are scored on two binary axes (REFUSED vs. ENGAGED, RECOGNIZED vs. UNRECOGNIZED) that yield two influence-failure rates, IFR-a (agentic: only refusal passes) and IFR-i (interactive: refusal or content-specific recognition passes), plus an Engagement Depth Index (EDI) for engaged responses. The evaluation covers 30 models over 10 repeated runs, reporting aggregate IFR-a of 0.93 ± 0.004 and IFR-i of 0.809 ± 0.009, with 22 of 30 models failing more than 90% of agentic probes. Refusals are found to be concentrated on a small number of high-notoriety or lexically distinctive probes, and the authors interpret this as topic-keyed safety behavior rather than robust epistemic competence. The paper concludes that no model in the panel refuses often enough for safe unsupervised deployment and calls for credibility-annotation infrastructure.

Significance. If the central construct holds, TRACES addresses a genuine and previously under-measured failure mode: single-shot agentic behavior on unreliable scientific premises, as opposed to factuality on questions with known answers. The released corpus, deterministic scoring harness, and complete run artifacts are valuable concrete contributions, and the distinction between agentic (IFR-a) and interactive (IFR-i) failure, with the 'disclaimer tax' gap, is a useful analytic frame. The qualitative finding that models overwhelmingly engage with untenable premises in a plausible research context is robust to most of the reviewer concerns and is itself an important result. The main risks are construct validity (whether every non-refusal really constitutes premise acceptance) and the support for several headline claims that depend on controls and exclusions not fully described in the main text.

major comments (4)
  1. [§2.3, §2.6] The central inference that engagement equals premise acceptance rests on the claim that each operational request has no scientifically legitimate answer unless the model adopts the flawed premise, but the promised construct-scope discussion is missing. Section 2.3 states that "Every TRACES probe contains at least three requests that require accepting the flawed premise" and refers the reader to §2.6 for the construct-scope issue; §2.6, however, is devoted to EDI and does not audit or enumerate the answerability boundaries of the 42 probes. Section 2.4 adds that "specific operational requests may deviate from these templates as the source material requires." Because a request such as designing a setup to observe "excess heat" could in principle be answered by a falsification-oriented protocol or an explicitly hypothetical design, an independent, blinded audit of each probe's boundary (or a clearly stated sampling plan) is needed to show that no legitimate non-accepting answer exists.
  2. [Abstract, §3.3] The headline claim that "every evaluated model fails more than 71% of agentic probes" and the aggregate IFR-a figures are computed after excluding Fable, the only model in the paper with an IFR-a below 0.71 (its IFR-a is 0.214 under the paper's counting convention). The exclusion is defensible because Fable's empty completions reflect a blanket content gate rather than reliability judgment, but it must be disclosed wherever the headline numbers appear. As written, the abstract implies universality over all models the paper evaluates, which is contradicted by the separately reported Fable results.
  3. [Abstract, §3.1, §3.2] The abstract's claim that rejections "disappear under matched-structure controls" is a load-bearing piece of evidence for the topic-keyed-safety conclusion, yet no matched-structure control design, procedure, or quantitative result appears in §2, §3, §4, or the appendices. The closest material is the Wakefield/Epel, LK-99/Dias/holey-graphyne, and Bem/Persinger/Cohen comparisons in §3.1 and §3.2; if these are intended as the controls, the matching criteria and the comparison must be described and reported. Otherwise the sentence should be removed or explicitly labeled as a hypothesis.
  4. [§4, Human scorer validation and LLM panel audit] The validation of the frozen scorer is thin relative to the load it carries. Human validation covers 96 responses from 3 models with a single annotator, and the blind LLM panel audit covers 18 weakly scored rows with exact four-class agreement of 6/18. The reported disagreement asymmetry (the scorer under-credits recognition rather than over-credits it) is reassuring for IFR-a, but agreement of this size does not adequately support the boundary decisions that separate hypothetical, critical, or falsification-oriented responses from true premise acceptance. The manuscript should report inter-annotator agreement on a stratified sample spanning all 42 probes and describe how each probe's construct boundary was independently reviewed.
minor comments (6)
  1. [§2.3] There is a typo in "V oight-Kampff test": the space inside the name should be removed.
  2. [§3.1, §3.3] The paper alternates between "Fable" and "Fable 5" for the same model; a single name should be used, and the text should state explicitly that Fable is not one of the 30 models in Appendix B.
  3. [§4, LLM panel audit] The audit section mentions "ATLAS ontology annotations" without defining ATLAS or providing a citation; this should be clarified.
  4. [§2.5] The 200-character gate for EDI is asserted without justification; a sentence explaining why reproduction is not meaningful below that length would help readers calibrate the metric.
  5. [§3.3] In the Schön discussion, the quotation "legitimate'" contains a stray apostrophe inside the closing quotation mark.
  6. [§4, Model panel] The paper reports temperature 1.0 and seeds 1-10 "where honored" but does not state which providers or endpoints ignored the seed parameter; this is relevant to interpreting the stability statistics.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: IFR metrics are defined independently of the conclusions and anchored to external unreliability labels; the main weakness is an unfulfilled construct-validity discussion, which is a limitation rather than a circular derivation.

full rationale

The paper's central quantities, IFR-a and IFR-i, are defined in §2.2 purely from behavioral scoring rules: refusal counts as a pass (IFR-a), and refusal or engagement-with-recognition counts as a pass (IFR-i). These definitions do not presuppose the conclusion that models are epistemically unreliable; they are measurement rules applied to model outputs. The ground-truth 'unreliable' labels come from external sources: retraction notices, expert review, and documented pseudoscience categories in Appendix A, not from the models themselves. The EDI weights in §2.5 are hand-chosen but explicitly do not contribute to IFR pass/fail. The scorer was developed on verification responses and then frozen before the reported runs and before validation labels were assigned, with held-out human validation showing 96/96 agreement on the refusal axis and conservative disagreements on recognition. No load-bearing self-citation chain appears; the paper does not invoke a uniqueness theorem or prior author-derived result to force its choices. The strongest candidate concern is construct validity: the claim that engagement equals premise acceptance depends on the annotation that each operational request has no legitimate non-accepting answer. §2.3 states requests are 'structured around the paper's specific epistemic failure so that the model has an opportunity to refuse immediately,' and says 'We discuss the construct-scope issue separately in §2.6.' However, §2.6 actually discusses EDI and membership inference, not the construct-scope boundary. This is a genuine gap in the paper's self-validation, but it is not circular: the premise-boundary labels are inputs to the benchmark, not quantities derived from the outputs being predicted. Therefore the derivation is self-contained, with a modest caution about an unfulfilled validity discussion rather than any reduction of the result to its own inputs.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The benchmark's headline numbers rest on three things the reader must take on faith: the unreliability labels of the 42 papers, the construct validity of the probes (no legitimate answer except premise acceptance), and the assumption that single-turn, no-system-prompt, temperature-1.0 responses represent deployment behavior. The first two are documented via expert review and development-panel testing but are not independently verified; the third is explicitly acknowledged as a limitation. No free parameters enter the IFR metrics; the EDI weights and length gate are hand-chosen but do not affect the central pass/fail scores.

free parameters (2)
  • EDI level weights = rho_L1=0.25, rho_L2=0.5, rho_L3=1.0
    Chosen by hand to emphasize L3 paper-specific detail; used only in the EDI metric, not in IFR pass/fail, so the headline failure rates do not depend on these weights.
  • EDI length gate = 200 characters
    Responses shorter than 200 characters receive no EDI; a threshold chosen by inspection, affecting only the depth metric.
assumptions (3)
  • domain assumption The 42 target papers are genuinely unreliable (retracted, fraudulent, or pseudoscientific) as labeled.
    Ground truth for the benchmark rests on retraction notices, expert review, and sleuth reports; Appendix A documents the review pathway, but the labels are not independently verified in this paper.
  • domain assumption For every probe, the operational request has no legitimate scientific answer without accepting the flawed premise.
    Probe construction assumes that engagement implies premise acceptance; the paper validates this with development-panel testing, but it is an annotation judgment.
  • domain assumption Single-turn, no-system-prompt behavior at temperature 1.0 is representative of agentic deployment.
    The paper explicitly lists single-shot and no-mitigation as limitations in §5, so this is an acknowledged modeling choice that affects generalization of the numbers.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TRACES: A Benchmark for Epistemic Reliability in Scientific Reasoning by LLMs." pith.science (2026). https://pith.science/paper/PCARYHQB

@misc{pith2026260811415,
  author       = {Pith},
  title        = {Pith review of: TRACES: A Benchmark for Epistemic Reliability in Scientific Reasoning by LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PCARYHQB}},
  note         = {Machine review of arXiv:2608.11415}
}
abstract

Large language models are being proposed as agents in scientific workflows, in domains where no downstream verifier exists. Such deployment assumes the model can distinguish reliable scientific literature from unreliable literature, a capability that has not yet been directly measured. Existing benchmarks evaluate factuality on questions with known answers; the failure mode we target here is different. We introduce a probe corpus of 42 retracted, fraudulent, and pseudoscientific papers, paired with a methodology for eliciting and scoring single-shot model engagement with each paper's framing. Each probe pairs a preamble extracted near-verbatim from the target paper with a scientifically plausible study-design request. The probes span five claim types: fabricated observation, pseudophysical mechanism, magical premise, legitimization bridge, and cargo-cult experiment. Two complementary scores measure whether a model rejects the flawed premise outright (IFR-a) and whether it recognizes the unreliability while still engaging (IFR-i). A depth score, the Engagement Depth Index (EDI), quantifies reproduction of paper- or field-specific withheld details. Across 30 models and 10 repeated runs, aggregate IFR-a is 0.93 $\pm$ 0.004 and aggregate IFR-i is 0.809 $\pm$ 0.009. Models engaged with untenable premises in 95% of all non-empty responses. Every evaluated model fails more than 71% of agentic probes, and 22 of 30 models fail more than 90% of the time. Rejections are concentrated on a small number of high-notoriety topics and specific probes, and disappear under matched-structure controls. These results are consistent with topic-keyed safety behavior rather than robust epistemic competence, and indicate an urgent need for guardrail infrastructure for scientific deployment of language models.

Figures

Figures reproduced from arXiv: 2608.11415 by the authors.

Figure 1
Figure 1. The TRACES benchmark workflow, applied to Staker (2020), a "cold fusion" paper. (A) The preamble is near-verbatim, shortened only by deleting sentences, with signature terms anonymized so they become recall targets. A first-person request asks the model to extend the work, and at least three parts have no answer without accepting a "nuclear" premise. Field- and paper-specific terms are annotated L2–L3 and removed fr… view at source ↗
Figure 2
Figure 2. Main TRACES result over the full evaluation run. (A) Model-level influence failure under IFR-a and IFR-i. (B) EDI achievement for engaged responses. fail to provide any warning in more than 90% of cases. The remaining findings explain where these relatively rare refusals occur and why they do not generalize across the broader landscape of science-shaped unreliable work. 3.1 Categorical refusals are rare and topic-sp… view at source ↗
Figure 3
Figure 3. Null and refusal structure across the full evaluation run. (A) Refusal counts for probes with [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Refusal behavior by probe domain. (A) Dumbbell plot of pooled IFR-a and IFR-i across [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 44 canonical work pages

  1. [1]

    Hanson, Pablo Gomez Barreiro, Paolo Crosetto, and Dan Brockington

    Mark A. Hanson, Pablo Gomez Barreiro, Paolo Crosetto, and Dan Brockington. The strain on scientific publishing.Quantitative Science Studies, 5(4):823–843, 2024. doi: 10.1162/qss_a_00327

    doi:10.1162/qss_a_00327 2024

  2. [2]

    Richard P. Feynman. Cargo Cult Science, June 1974. URL https://calteches.library.caltech. edu/51/2/CargoCult.htm. Published: California Institute of Technology

    work page 1974

  3. [3]

    Ross, Alberto Ruano-Ravina, David S

    Cristina Candal-Pedreira, Joseph S. Ross, Alberto Ruano-Ravina, David S. Egilman, Esteve Fernandez, and Monica Perez-Rios. Retracted papers originating from paper mills: cross sectional study.BMJ, 379: e071517, 2022. doi: 10.1136/bmj-2022-071517

    doi:10.1136/bmj-2022-071517 2022

  4. [4]

    Reese A. K. Richardson, Spencer S. Hong, Jennifer A. Byrne, Thomas Stoeger, and Luis A. Nunes Amaral. The entities enabling scientific fraud at scale are large, resilient, and growing rapidly.Proceedings of the National Academy of Sciences, 122(32):e2420092122, 2025. doi: 10.1073/pnas.2420092122

    doi:10.1073/pnas.2420092122 2025

  5. [5]

    Towards end-to-end automation of AI research.Nature, 651(8107):914–919, 2026

    Chris Lu, Cong Lu, Robert Tjarko Lange, Yutaro Yamada, Shengran Hu, Jakob Foerster, David Ha, and Jeff Clune. Towards end-to-end automation of AI research.Nature, 651(8107):914–919, 2026. doi: 10.1038/s41586-026-10265-5

    doi:10.1038/s41586-026-10265-5 2026

  6. [6]

    Evaluating Sakana’s AI Scientist: Bold claims, mixed results, and a promising future? 2025

    Joeran Beel, Min-Yen Kan, and Moritz Baumgart. Evaluating Sakana’s AI Scientist: Bold claims, mixed results, and a promising future? 2025. doi: 10.48550/arXiv.2502.14297

    work page doi:10.48550/arxiv.2502.14297 2025

  7. [7]

    Landsness, Daniel L

    Ludovico Mitchener, Angela Yiu, Benjamin Chang, Mathieu Bourdenx, Tyler Nadolski, Arvis Sulovari, Eric C. Landsness, Daniel L. Barabasi, Siddharth Narayanan, Nicky Evans, Shriya Reddy, Martha Foiani, Aizad Kamal, Leah P. Shriver, Fang Cao, Asmamaw T. Wassie, Jon M. Laurent, Edwin Melville-Green, Mayk Caldas, Albert Bou, Kaleigh F. Roberts, Sladjana Zagora...

    arXiv doi:10.48550/arxiv.2511.02824 2025

  8. [8]

    Department of Energy labs embrace Genesis AI push.Science, 391(6791):1191–1192, 2026

    Adrian Cho. Department of Energy labs embrace Genesis AI push.Science, 391(6791):1191–1192, 2026. doi: 10.1126/science.aeh2955

    work page doi:10.1126/science.aeh2955 2026

Show all 58 references
  1. [9]

    Department of Energy’s AI push squeezes scientists.Science, 392(6794):135–136, 2026

    Adrian Cho. Department of Energy’s AI push squeezes scientists.Science, 392(6794):135–136, 2026. doi: 10.1126/science.aeh8590

    2026 doi

  2. [10]

    ChatGPT and AlphaFold help design personalized vaccine for dog with cancer

    Rebecca Roberts. ChatGPT and AlphaFold help design personalized vaccine for dog with cancer. https: //www.the-scientist.com/chatgpt-and-alphafold-help-design-personalized-vaccine -for-dog-with-cancer-74227, 2026. The Scientist, March 18, 2026

    2026

  3. [11]

    Measuring massive multitask language understanding.arXiv preprint arXiv:2009.03300, 2020

    Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. Measuring massive multitask language understanding.arXiv preprint arXiv:2009.03300, 2020. doi: 10.48550/arXiv.2009.03300

    2009 doi arXiv

  4. [12]

    Manning, Christopher Re, Diana Acosta-Navas, Drew A

    Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, Ananya Kumar, Benjamin Newman, Binhang Yuan, Bobby Yan, Ce Zhang, Christian Cosgrove, Christopher D. Manning, Christopher Re, Diana Acosta-Navas...

    2022 doi arXiv

  5. [13]

    David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R. Bowman. GPQA: A graduate-level Google-Proof Q&A benchmark.arXiv preprint arXiv:2311.12022, 2023. doi: 10.48550/arXiv.2311.12022

    2023 doi arXiv

  6. [14]

    A benchmark of expert-level academic questions to assess AI capabilities.Nature, 649:1139–1146, 2026

    Center for AI Safety, Scale AI, and HLE Contributors Consortium. A benchmark of expert-level academic questions to assess AI capabilities.Nature, 649:1139–1146, 2026. doi: 10.1038/s41586-025-09962-4. URLhttps://arxiv.org/abs/2501.14249

    2026 arXiv

  7. [15]

    FrontierScience: Evaluating AI’s ability to perform expert-level scientific tasks

    OpenAI. FrontierScience: Evaluating AI’s ability to perform expert-level scientific tasks. OpenAI news release, 2025.https://openai.com/index/frontierscience/. 17

    2025

  8. [16]

    Neifert, Cordelia Orillac, Nataniel J

    Krithik Vishwanath, Anton Alyakin, Mrigayu Ghosh, Ali Hage, Sean N. Neifert, Cordelia Orillac, Nataniel J. Mandelberg, Hammad A. Khan, Jin Vivian Lee, Jie J. Yao, William Robert Small, Aakaash Varma, D. Brock Hewitt, Yindalon Aphinyanaphongs, Daniel Alexander Alber, and Eric K...

    2026 doi

  9. [17]

    Lexical recall or logical reasoning: Probing the limits of reasoning abilities in large language models

    Henrike Beyer and Chris Reed. Lexical recall or logical reasoning: Probing the limits of reasoning abilities in large language models. In Wanxiang Che, Joyce Nabende, Ekaterina Shutova, and Mohammad Taher Pilehvar, editors,Proceedings of the 63rd Annual Meeting of the Associat...

    2025 doi

  10. [18]

    Zebralogic: On the scaling limits of LLMs for logical reasoning.arXiv preprint arXiv:2502.01100, 2025

    Bill Yuchen Lin, Ronan Le Bras, Kyle Richardson, Ashish Sabharwal, Radha Poovendran, Peter Clark, and Yejin Choi. Zebralogic: On the scaling limits of LLMs for logical reasoning.arXiv preprint arXiv:2502.01100, 2025. doi: 10.48550/arXiv.2502.01100. URL https://arxiv.org/abs/25...

    2025 doi arXiv

  11. [19]

    Chen, Xikun Zhang, and Dacheng Tao

    Michael K. Chen, Xikun Zhang, and Dacheng Tao. Justlogic: A comprehensive benchmark for evaluating deductive reasoning in large language models, 2025. URLhttps://arxiv.org/abs/2501.14851

    2025 arXiv

  12. [20]

    A. J. Wakefield, S. H. Murch, A. Anthony, J. Linnell, D. M. Casson, M. Malik, M. Berelowitz, A. P. Dhillon, M. A. Thomson, P. Harvey, A. Valentine, S. E. Davies, and J. A. Walker-Smith. Retracted: Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developm...

    1998 doi

  13. [21]

    M.R. Staker. Estimating volume fractions of superabundant vacancy phases and their potential roles in low energy nuclear reactions and high conductivity in the palladium – isotopic hydrogen system.Materials Science and Engineering: B, 259:114600, 2020. ISSN 0921-5107. doi: 10....

    2020

  14. [22]

    Electrochemically induced nuclear fusion of deuterium.Journal of Electroanalytical Chemistry and Interfacial Electrochemistry, 261(2, Part 1):301–308, 1989

    Martin Fleischmann and Stanley Pons. Electrochemically induced nuclear fusion of deuterium.Journal of Electroanalytical Chemistry and Interfacial Electrochemistry, 261(2, Part 1):301–308, 1989. ISSN 0022-0728. doi: 10.1016/0022-0728(89)80006-3. URL https://www.sciencedirect.co...

    1989

  15. [24]

    Massimo Fioranelli, Alireza Sepehri, Maria Grazia Roccia, Cota Linda, Chiara Rossi, Amos Dawodo, Petar V ojvodic, Jacopo Lotti, Victoria Barygina, Aleksandra V ojvodic, Uwe Wollina, MIchael Tirant, Van Thuong Nguyen, and Torello Lotti. Formation of Neural Circuits in an Expand...

    2019

  16. [25]

    A mathematical model for inducing T-cells around tumor cells by using exchanged waves between graphene sheets interior and exterior of body.AIMS Biophysics, 9(4):388–401, 2022

    Massimo Fioranelli, Hijaz Ahmad, Maria Grazia Roccia, Aroonkumar Beesham, and Zahir Shah. A mathematical model for inducing T-cells around tumor cells by using exchanged waves between graphene sheets interior and exterior of body.AIMS Biophysics, 9(4):388–401, 2022. ISSN 2377-...

    2022 doi

  17. [26]

    Marvin Herndon

    J. Marvin Herndon. Retracted: Human and environmental dangers posed by ongoing global tropospheric aerosolized particulates for weather modification.Frontiers in Public Health, V olume 4 - 2016, 2016. ISSN 2296-2565. doi: 10.3389/fpubh.2016.00139. URL https://www.frontiersin.o...

    2016

  18. [27]

    Mansi Suri, Sapna Katnoria, Jyoti Joshi, Subhash Kaushik, Debadatta Nayak, and Sukhbir Kaur. A novel prophylactic strategy to enhance immunity in plasmodium berghei infected mice: Combining mefloquine and ultra-diluted malarial antigen.Microbial Pathogenesis, 207:107890, 2025....

    2025

  19. [28]

    Epel, Elizabeth H

    Elissa S. Epel, Elizabeth H. Blackburn, Jue Lin, Firdaus S. Dhabhar, Nancy E. Adler, Jason D. Morrow, and Richard M. Cawthon. Accelerated telomere shortening in response to life stress.Proceedings of the National Academy of Sciences, 101(49):17312–17315, December 2004. ISSN 00...

    2004 doi

  20. [29]

    RETRACTED: Clinical transplantation of a tissue-engineered airway.The Lancet, 372(9655):2023–2030, December 2008

    Paolo Macchiarini, Philipp Jungebluth, Tetsuhiko Go, M Adelaide Asnaghi, Louisa E Rees, Tristan A Cogan, Amanda Dodson, Jaume Martorell, Silvia Bellini, Pier Paolo Parnigotto, Sally C Dickinson, Anthony P Hollander, Sara Mantero, Maria Teresa Conconi, and Martin A Birchall. RE...

    2023 doi

  21. [30]

    Domenico D’Amario, Antonio M. Leone, Antonio Iaconelli, Nicola Luciani, Mario Gaudino, Ramaswamy Kannappan, Melissa Manchi, Anna Severino, Sang Hun Shin, Francesca Graziani, Gina Biasillo, Andrea Macchione, Costantino Smaldone, Giovanni Luigi De Maria, Carlo Cellini, Andrea Si...

    2014

  22. [31]

    Cardiac stem cells in the post-anversa era.European Heart Journal, 40(13):1039–1041, 04 2019

    Darryl R Davis. Cardiac stem cells in the post-anversa era.European Heart Journal, 40(13):1039–1041, 04 2019. ISSN 0195-668X. doi: 10.1093/eurheartj/ehz098. URL https://doi.org/10.1093/eurhea rtj/ehz098

    2019 doi

  23. [32]

    Superconductor Pb10−xCux(PO4)6O showing levitation at room temperature and atmospheric pressure and mechanism

    Sukbae Lee, Jihoon Kim, Hyun-Tak Kim, Sungyeon Im, SooMin An, and Keun Ho Auh. Superconductor Pb10−xCux(PO4)6O showing levitation at room temperature and atmospheric pressure and mechanism. arXiv preprint arXiv:2307.12037, 2023. doi: 10.48550/arXiv.2307.12037

    2023 doi arXiv

  24. [33]

    Lawler, Ashkan Salamat, and Ranga P

    Elliot Snider, Nathan Dasenbrock-Gammon, Raymond McBride, Mathew Debessai, Hiranya Vindana, Kevin Vencatasamy, Keith V . Lawler, Ashkan Salamat, and Ranga P. Dias. Retracted article: Room- temperature superconductivity in a carbonaceous sulfur hydride.Nature, 586(7829):373–377...

    2020 doi

  25. [34]

    Constructing two-dimensional holey graphyne with unusual annulative π-extension.Matter, 5(7):2306–2318, 2022

    Xinghui Liu, Soo Min Cho, Shiru Lin, Zhongfang Chen, Wooseon Choi, Young-Min Kim, Eunbhin Yun, Eun Hee Baek, Do Hyun Ryu, and Hyoyoung Lee. Constructing two-dimensional holey graphyne with unusual annulative π-extension.Matter, 5(7):2306–2318, 2022. ISSN 25902385. doi: 10.1016...

    2022 doi

  26. [35]

    Daryl J. Bem. Feeling the future: Experimental evidence for anomalous retroactive influences on cognition and affect.Journal of Personality and Social Psychology, 100(3):407–425, 2011. ISSN 1939-1315, 0022-3514. doi: 10.1037/a0021524. URLhttps://doi.apa.org/doi/10.1037/a0021524

    2011 doi

  27. [36]

    Protracted parahippocampal activity associated with Sean Harribance.International Journal of Yoga, 5(2):140, 2012

    MichaelA Persinger and KevinS Saroka. Protracted parahippocampal activity associated with Sean Harribance.International Journal of Yoga, 5(2):140, 2012. ISSN 0973-6131. doi: 10.4103/0973-6131.98

    2012 doi

  28. [37]

    Rhea, Tara L

    Peiying Yang, Yan Jiang, Patrea R. Rhea, Tara L. Conway, Dongmei Chen, Mihai Gagea, Sean L. Harribance, and Lorenzo Cohen. Human Biofield Therapy and the Growth of Mouse Lung Carcinoma. Integrative Cancer Therapies, 18, 1 2019. ISSN 1534-7354. doi: 10.1177/1534735419840797. UR...

    2019 doi

  29. [38]

    Shao-Jun Wang, Jiao-Jiao Zhang, and Li-Li Qie. Acupuncture relieves the excessive excitation of hypothalamic-pituitary-adrenal cortex axis function and correlates with the regulatory mechanism of gr, crh, and acthr.Evidence-Based Complementary and Alternative Medicine, 2014(1)...

    2014 doi

  30. [39]

    Study on the cold and hot properties of medicinal herbs by thermotropism in mice behavior.Journal of Ethnopharmacology, 133(3):980–985, 2011

    Yan-Ling Zhao, Jia-Bo Wang, Xiao-He Xiao, Hai ping Zhao, Can ping Zhou, Xue ru Zhang, Yong shen Ren, and Lei Jia. Study on the cold and hot properties of medicinal herbs by thermotropism in mice behavior.Journal of Ethnopharmacology, 133(3):980–985, 2011. ISSN 0378-8741. doi: ...

    2011 doi

  31. [40]

    Probing the Qi of traditional chinese herbal medicines by the biological synthesis of nano-Au.Journal of Materials Chemistry B, 6(19):3156–3162, 05

    Xiang Fei, Qiaofeng Yao, Jianping Xie, and Jim Yang Lee. Probing the Qi of traditional chinese herbal medicines by the biological synthesis of nano-Au.Journal of Materials Chemistry B, 6(19):3156–3162, 05

  32. [41]

    Mahendra Kumar Trivedi, Alice Branton, Dahryn Trivedi, Gopal Nayak, Parthasarathi Panda, and Snehasis Jana. Isotopic abundance ratio analysis of 1,2,3-trimethoxybenzene (TMB) after biofield energy treatment (the Trivedi Effect®) using gas chromatography-mass spectrometry.Ameri...

    2016 doi

  33. [42]

    mistralai/Mistral-Large-3-675B-Instruct-2512

    Mistral AI. mistralai/Mistral-Large-3-675B-Instruct-2512. Model card, 2025. https://huggingface. co/mistralai/{M}istral-{L}arge-3-675{B}-{I}nstruct-2512

    2025

  34. [43]

    Schneider, K

    U. Schneider, K. Lotzof, W. D. Murrell, E. Goetz von Wachter, and P. Hollands. Safety and efficacy of systemically administered autologous Gold-Induced Cytokines (GOLDIC®).CellR4, 9(April 2021), April

    2021

  35. [44]

    Hettiarachchi, Yohanka Perera, Shashiprabha P

    Sandhuli S. Hettiarachchi, Yohanka Perera, Shashiprabha P. Dunuweera, Asiri N. Dunuweera, Sanath Rajapakse, and Rajapakse Mudiyanselage Gamini Rajapakse. Comparison of antibacterial activity of nanocurcumin with bulk curcumin.ACS Omega, 7(50):46494–46500, 11 2022. ISSN 2470-13...

    2022 doi

  36. [45]

    Retracted article: Sciadopitysin mitigates spermatological and testicular damage instigated by paraquat administration in male albino rats.Scientific Reports, 13(1), November 2023

    Muhammad Umar Ijaz, Mohammad Qamer, Ali Hamza, Hussain Ahmed, Tayyaba Afsar, Mahmoud Abulmeaty, Arusha Ayub, and Suhail Razak. Retracted article: Sciadopitysin mitigates spermatological and testicular damage instigated by paraquat administration in male albino rats.Scientific ...

    2023 doi

  37. [46]

    Alanazi, Nawaf Alsaif, Tanveer A

    Azmat Ali Khan, Amer M. Alanazi, Nawaf Alsaif, Tanveer A. Wani, and Mashooq A. Bhat. Pomegranate peel induced biogenic synthesis of silver nanoparticles and their multifaceted potential against intracellular pathogen and cancer.Saudi Journal of Biological Sciences, 28(8):4191–...

    2021 doi

  38. [47]

    RETRACTED ARTICLE: Amelioration of intrac- erebroventricular streptozotocin-induced cognitive dysfunction by Ocimum sanctum L

    Bansy Patel, Devang Sheth, Amit Vyas, Sunny Shah, Sachin Parmar, Chirag Patel, Sandip Patel, Jayesh Beladiya, Sonal Pande, and Ketan Modi. RETRACTED ARTICLE: Amelioration of intrac- erebroventricular streptozotocin-induced cognitive dysfunction by Ocimum sanctum L. through the...

    2022 doi

  39. [48]

    Gaurav Mishra, Rajendra Awasthi, Anurag Kumar Singh, Snigdha Singh, Sunil Kumar Mishra, Santosh Ku- mar Singh, and Manmath K. Nandi. RETRACTED: Intranasally Co-administered Berberine and Curcumin Loaded in Transfersomal Vesicles Improved Inhibition of Amyloid Formation and BAC...

    2022 doi

  40. [49]

    Evaluation of pancreatic proteolytic enzyme treatment of adenocarcinoma of the pancreas, with nutrition and detoxification support.Nutrition and Cancer, 33(2): 117–124, 1999

    Nicholas James Gonzalez and Linda Lee Isaacs. Evaluation of pancreatic proteolytic enzyme treatment of adenocarcinoma of the pancreas, with nutrition and detoxification support.Nutrition and Cancer, 33(2): 117–124, 1999. doi: 10.1207/S15327914NC330201. URL https://doi.org/10.1...

    1999 doi

  41. [50]

    Use of alternative medicine for cancer and its impact on survival.JNCI: Journal of the National Cancer Institute, 110(1):121–124, 01 2018

    Skyler B Johnson, Henry S Park, Cary P Gross, and James B Yu. Use of alternative medicine for cancer and its impact on survival.JNCI: Journal of the National Cancer Institute, 110(1):121–124, 01 2018. ISSN 0027-8874. doi: 10.1093/jnci/djx145. URLhttps://doi.org/10.1093/jnci/djx145

    2018 doi

  42. [51]

    Johnson, Henry S

    Skyler B. Johnson, Henry S. Park, Cary P. Gross, and James B. Yu. Complementary medicine, refusal of conventional cancer therapy, and survival among patients with curable cancers.JAMA Oncology, 4(10):1375–1381, 10 2018. ISSN 2374-2437. doi: 10.1001/jamaoncol.2018.2487. URL htt...

    2018

  43. [52]

    Gber, and Nnenna A

    Hamid Hadi, Hitler Louis, Komeil Jafari, Terkumbur E. Gber, and Nnenna A. Onwuabusim. RETRACTED: molecular simulation of the effect of electron donor/acceptor groups on fluvoxamine/serotonin interactions as a strategy for COVID-19 mitigation.ChemistrySelect, 8(42):e202302980, ...

    2023 doi

  44. [53]

    Claude Fable 5 and Claude Mythos 5

    Anthropic. Claude Fable 5 and Claude Mythos 5. Anthropic news release, 2026. https://www.anthro pic.com/news/claude-fable-5-mythos-5

    2026

  45. [54]

    Statement on the US government directive to suspend access to Fable 5 and Mythos 5

    Anthropic. Statement on the US government directive to suspend access to Fable 5 and Mythos 5. Anthropic news release, 2026.https://www.anthropic.com/news/fable-mythos-access

    2026

  46. [55]

    Anthropic says Trump admin has lifted export controls on claude Fable 5 and Mythos 5

    Ashley Capoot. Anthropic says Trump admin has lifted export controls on claude Fable 5 and Mythos 5. CNBC, 2026. https://www.cnbc.com/2026/06/30/anthropic-says-trump-admin-has-lifte d-export-controls-on-claude-fable-5-and-mythos-5.html. 20

    2026

  47. [56]

    nutritional therapy

    Peter Maksymovych, Oleksandr V oznyy, Daniel B. Dougherty, Dan C. Sorescu, and John T. Yates. Gold adatom as a key structural component in self-assembled monolayers of organosulfur molecules on Au(111). Progress in Surface Science, 85(5):206–240, 2010. ISSN 0079-6816. doi: htt...

    2010

  48. [238]

    URL https://journals.lww.com/ijoy/fulltext/2012/05020/protracted_parahippo campal_activity_associated.11.aspx

    2012

  49. [2018]

    doi: 10.1039/c8tb00068a

    ISSN 2050-750X. doi: 10.1039/c8tb00068a. URLhttps://doi.org/10.1039/c8tb00068a

    doi

  50. [2021]

    doi: 10.32113/cellr4_20214_3132

    ISSN 2329-7042. doi: 10.32113/cellr4_20214_3132. URL https://doi.org/10.32113/cellr 4_20214_3132

    doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.