{"paper":{"title":"Anchor and Check: Deterministic Cross-Prompt, Cross-Model Reconciliation for Auditable LLM Decisions","arxiv_id":"2611.25818","primary_cat":"cs.AI","submitted_at":"2026-11-19T09:38:16+00:00","version_compiled":"v1"},"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 v1","sentences":[{"i":0,"p":0.99,"id":"s0","own":false,"role":"background","text":"Pipelines that let a large language model (LLM) make discrete decisions need to know which decisions they may act on unattended, and a record that explains each one.","hedge":{"p":0.97,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.01,"method":0.0,"result":0.0,"background":0.99,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.87,"id":"s1","own":true,"role":"method","text":"We study a small, fully specified reconciliation contract: commit the target model's greedy answer (the anchor), draw one or two greedy checks that reach the answer by a different route (an alternate prompt, or a model from another family), and let a pure function return one of five outcomes, of which only Agree is acted on.","hedge":{"p":0.77,"label":"asserted"},"own_p":0.73,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.82,"role_probabilities":{"gap":0.0,"method":0.87,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.13,"interpretation":0.0}},{"i":2,"p":1.0,"id":"s2","own":true,"role":"result","text":"On 48,000 model-item evaluations over 6,000 questions, with every method given access to verbalized confidence as a tie-breaker, the main finding concerns where calls come from: pooling another family's samples raises semantic entropy's AUROC by 1.7 to 4.0 points at every budget, and a same-model temperature check is the weakest design.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.85,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"result","text":"As complete pipelines, the point estimates of the three-call contract and pooled semantic entropy over 3 or 5 samples differ by less than one point, with intervals containing zero.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.06,"quantitative":true,"quantitative_p":0.98,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"result","text":"As detectors of the anchor's errors, the contract is ahead of pooled entropy with 3 and 5 samples under both tie-breaker conventions; with 10 samples the point estimates differ by less than one point and the sign depends on the convention.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.98,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"Under the strict Agree gate it automates 70% of items and still acts on 16% of errors.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.07,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":6,"p":0.5,"id":"s6","own":true,"role":"limitation","text":"What the contract adds is a fixed committed answer, an explicit outcome and a decision replayable from its record; generation itself was not reproducible on the hosted API.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.83,"main_claim_p":0.1,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.02,"background":0.0,"limitation":0.5,"future_work":0.0,"contribution":0.28,"interpretation":0.2}}]},"compiled":{"main_claim":{"p":0.99,"text":"On 48,000 model-item evaluations over 6,000 questions, with every method given access to verbalized confidence as a tie-breaker, the main finding concerns where calls come from: pooling another family's samples raises semantic entropy's AUROC by 1.7 to 4.0 points at every budget, and a same-model temperature check is the weakest design.","location":"abstract s2","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":5944,"cost_usd_reported":0.0},"compiled_at":"2026-11-20","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.04,"states_limitation":0.95,"human_participants":0.05,"controlled_comparison":0.82,"randomized_assignment":0.03,"states_quantitative_result":0.99,"compares_with_prior_results":0.94,"mentions_code_or_data_availability":0.05},"evidential_basis":{"p":0.95,"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":0.95,"simulation_or_computation":0.05}},"contribution_type":{"p":0.73,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":0.73,"empirical_study":0.27,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.97,"text":"We study a small, fully specified reconciliation contract: commit the target model's greedy answer (the anchor), draw one or two greedy checks that reach the answer by a different route (an alternate prompt, or a model from another family), and let a pure function return one of five outcomes, of which only Agree is acted on.","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 v1"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":7,"sentences_in_abstract":7},"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,"review_comments_judged":{"judged_by":["gpt-5.6-luna","grok-4.7"],"judged_against":"Pith's own text copy of each paper: text extracted from the PDF, the first 90,000 characters","measured_on":"2026-10-04","accurate":113,"decided":149,"sampled_reviews":200,"split_by_lane":false,"by_lane":"https://pith.science/accuracy#by-lane"}}}