{"paper":{"title":"CordisBench: Can Language Models Reason About Component Lifecycles in Dynamic Agent Harnesses?","arxiv_id":"2609.01600","primary_cat":"cs.CL","submitted_at":"2026-09-01T17:59:13+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":1.0,"id":"s0","own":false,"role":"background","text":"Dynamic agent harnesses let language models change the software that shapes their own execution.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.07,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.02,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.93,"id":"s1","own":false,"role":"background","text":"This flexibility brings a new reasoning burden: a local plugin change can propagate through dependencies and cleanup.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.09,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.07,"method":0.0,"result":0.0,"background":0.93,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.99,"id":"s2","own":true,"role":"contribution","text":"We introduce CordisBench, a 1,200-question benchmark of this lifecycle reasoning.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.04,"quantitative":true,"quantitative_p":0.98,"role_probabilities":{"gap":0.0,"method":0.01,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.99,"interpretation":0.0}},{"i":3,"p":0.72,"id":"s3","own":true,"role":"method","text":"It combines a controlled formal setting with programs executed against Cordis, a runtime that manages component dependencies and cleanup, and asks models to identify affected components, predict state after a specified teardown order, determine which conditions hold under all or some orders, and choose reconfigurations that succeed when executed.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.89,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.35,"role_probabilities":{"gap":0.0,"method":0.72,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.28,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"method","text":"Across these tasks, we evaluate three efficiency-oriented models at low reasoning effort with 2, 4, 8, 16, 24, or 32 relevant interactions, using deterministic task-specific scoring.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.94,"main_claim_p":0.04,"quantitative":true,"quantitative_p":0.97,"role_probabilities":{"gap":0.0,"method":1.0,"result":0.0,"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":"Models usually handle small systems well but grow less reliable as more interactions become relevant, especially when predicting final state and when reasoning across teardown orders.","hedge":{"p":0.95,"label":"asserted"},"own_p":0.94,"main_claim_p":0.6,"quantitative":true,"quantitative_p":0.54,"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":0.97,"id":"s6","own":true,"role":"result","text":"Additional inference effort recovers marked gains for some models.","hedge":{"p":0.87,"label":"asserted"},"own_p":0.9,"main_claim_p":0.12,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.97,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.03}},{"i":7,"p":0.64,"id":"s7","own":true,"role":"result","text":"The cost is nontrivial: on our 16-interaction subset, GPT-5.6 Luna uses nearly 3,000 reasoning tokens per question at medium effort.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.64,"background":0.0,"limitation":0.35,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":8,"p":0.81,"id":"s8","own":true,"role":"result","text":"For these controlled instances, that cost is avoidable: an independent finite reference semantics agrees with Cordis execution on every observation and action outcome used for scoring across all 528 executable questions.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.91,"main_claim_p":0.09,"quantitative":true,"quantitative_p":0.97,"role_probabilities":{"gap":0.0,"method":0.03,"result":0.81,"background":0.0,"limitation":0.02,"future_work":0.0,"contribution":0.02,"interpretation":0.12}}]},"compiled":{"main_claim":{"p":0.98,"text":"Models usually handle small systems well but grow less reliable as more interactions become relevant, especially when predicting final state and when reasoning across teardown orders.","location":"abstract s5","hedge_strength":{"p":0.95,"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":6925,"cost_usd_reported":0.0},"compiled_at":"2026-09-23","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.08,"states_limitation":0.92,"human_participants":0.04,"controlled_comparison":0.29,"randomized_assignment":0.02,"states_quantitative_result":0.98,"compares_with_prior_results":0.23,"mentions_code_or_data_availability":0.06},"evidential_basis":{"p":0.96,"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":0.96,"simulation_or_computation":0.04}},"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":0.99,"text":"We introduce CordisBench, a 1,200-question benchmark of this lifecycle reasoning.","location":"abstract s2"}},"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":9,"sentences_in_abstract":9},"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"}}}