{"paper":{"title":"PaperAudit-Bench: Benchmarking Error Detection in Research Papers for Critical Automated Peer Review","arxiv_id":"2601.19916","primary_cat":"cs.CL","submitted_at":"2026-01-07T04:26:12+00:00","version_compiled":"v2"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v2","sentences":[{"i":0,"p":0.68,"id":"s0","own":false,"role":"background","text":"Large language models (LLMs) can generate fluent peer reviews but often miss sparse, subtle, and document-level scientific errors.","hedge":{"p":0.91,"label":"asserted"},"own_p":0.12,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.32,"method":0.0,"result":0.0,"background":0.68,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":1.0,"id":"s1","own":true,"role":"contribution","text":"We introduce PaperAudit-Bench, a benchmark for evaluating paper-level error detection and error-aware automated review.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.92,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":1.0,"interpretation":0.0}},{"i":2,"p":0.6,"id":"s2","own":true,"role":"contribution","text":"It comprises (1) PaperAudit-Dataset, 220 high-quality AI/ML papers containing more than 2,200 controlled synthetic errors across eight categories, and (2) PaperAudit-Review, an evidence-aware framework that links structured error detection to review generation.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.9,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.97,"role_probabilities":{"gap":0.0,"method":0.4,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.6,"interpretation":0.0}},{"i":3,"p":0.99,"id":"s3","own":true,"role":"result","text":"Experiments across models, error categories, paper sections, and analysis depths show that sparse long-context auditing remains challenging for current LLMs.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.3,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.99,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":4,"p":0.94,"id":"s4","own":true,"role":"result","text":"Conditioning review generation on explicit, manuscript-grounded error evidence produces more conservative and discriminative assessments that align more closely with human judgments than representative automated-review baselines.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.91,"main_claim_p":0.55,"quantitative":true,"quantitative_p":0.88,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.94,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.06}},{"i":5,"p":0.93,"id":"s5","own":true,"role":"result","text":"The dataset also supports post-training compact detectors through SFT and RL; Qwen3-14B achieves 55.8% F1 at Detection@1 under a fixed held-out protocol.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.91,"main_claim_p":0.08,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.02,"result":0.93,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.05,"interpretation":0.0}},{"i":6,"p":0.59,"id":"s6","own":true,"role":"contribution","text":"A PubPeer transfer study and deployment in ScienceOne, a scientific foundation-model platform led by the Chinese Academy of Sciences, further illustrate practical applicability.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.82,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.05,"result":0.26,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.59,"interpretation":0.1}},{"i":7,"p":0.84,"id":"s7","own":false,"role":"contribution","text":"Code and Dataset Page: https://github.com/TU2021/PaperAudit ScienceOne Manuscript Review: https://www.scienceone.cn/lit/#/assistant","hedge":{"p":0.81,"label":"unclear"},"own_p":0.46,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.02,"result":0.0,"background":0.02,"limitation":0.0,"future_work":0.12,"contribution":0.84,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":0.86,"text":"Conditioning review generation on explicit, manuscript-grounded error evidence produces more conservative and discriminative assessments that align more closely with human judgments than representative automated-review baselines.","location":"abstract s4","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":6417,"cost_usd_reported":0.0},"compiled_at":"2026-10-20","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.07,"states_limitation":0.71,"human_participants":0.19,"controlled_comparison":0.9,"randomized_assignment":0.02,"states_quantitative_result":0.98,"compares_with_prior_results":0.93,"mentions_code_or_data_availability":0.98},"evidential_basis":{"p":0.99,"value":"observation or experiment","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.99,"simulation_or_computation":0.01}},"contribution_type":{"p":1.0,"value":"dataset or benchmark","probabilities":{"review":0.0,"method_or_tool":0.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":1.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.87,"text":"We introduce PaperAudit-Bench, a benchmark for evaluating paper-level error detection and error-aware automated review.","location":"abstract s1"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v2"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":8,"sentences_in_abstract":8},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448}}