{"paper":{"title":"HABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory","arxiv_id":"2610.24520","primary_cat":"cs.AI","submitted_at":"2026-10-17T07:30:09+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":"In long-term interactions with large language model (LLM) agents, users reveal habits through their choices and actions.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.08,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"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":1.0,"id":"s1","own":false,"role":"background","text":"These habits capture implicit user information that agents must infer across conversations.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.03,"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":2,"p":0.78,"id":"s2","own":false,"role":"background","text":"However, individual observations may provide only partial evidence about a habit or when it applies.","hedge":{"p":0.98,"label":"hedged_possibility"},"own_p":0.08,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.1,"role_probabilities":{"gap":0.22,"method":0.0,"result":0.0,"background":0.78,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":0.63,"id":"s3","own":false,"role":"gap","text":"Retaining past interactions alone does not ensure that agents can infer habits and apply them appropriately.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.11,"main_claim_p":0.15,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.63,"method":0.0,"result":0.0,"background":0.37,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"contribution","text":"We introduce HABIT-Bench to evaluate how well agent memory systems infer habits from weak evidence and apply them to later requests.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.91,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.04,"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":5,"p":0.99,"id":"s5","own":true,"role":"method","text":"We construct synthetic user histories from task-oriented dialogue data and controlled habit graphs.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.99,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":6,"p":0.99,"id":"s6","own":true,"role":"method","text":"The histories span hundreds of sessions across food, finance, software, and travel.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.9,"main_claim_p":0.02,"quantitative":true,"quantitative_p":0.82,"role_probabilities":{"gap":0.0,"method":0.99,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":7,"p":0.99,"id":"s7","own":true,"role":"method","text":"Four-choice tasks assess agents' ability to infer user habits and apply them appropriately.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.86,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.68,"role_probabilities":{"gap":0.0,"method":0.99,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":8,"p":0.98,"id":"s8","own":true,"role":"method","text":"They include temporary exceptions, changes in habits, and suggestions that users have not accepted.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.66,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.98,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.0}},{"i":9,"p":1.0,"id":"s9","own":true,"role":"method","text":"Sampled tasks undergo independent human review.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.81,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"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":10,"p":0.99,"id":"s10","own":true,"role":"method","text":"We evaluate eight memory methods with Qwen3-235B-A22B and GLM-5.3-Flash.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.95,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.89,"role_probabilities":{"gap":0.0,"method":0.99,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":11,"p":1.0,"id":"s11","own":true,"role":"result","text":"The best-performing memory method achieves 33.61\\% macro accuracy across the four domains.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.15,"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":12,"p":0.97,"id":"s12","own":true,"role":"result","text":"Performance remains limited even when annotated evidence is supplied directly.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.19,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.97,"background":0.0,"limitation":0.01,"future_work":0.0,"contribution":0.0,"interpretation":0.02}},{"i":13,"p":1.0,"id":"s13","own":true,"role":"interpretation","text":"These results expose a persistent gap between access to past interactions and the ability to infer and appropriately use user habits.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.88,"main_claim_p":0.87,"quantitative":false,"quantitative_p":0.13,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":1.0}}]},"compiled":{"main_claim":{"p":1.0,"text":"These results expose a persistent gap between access to past interactions and the ability to infer and appropriately use user habits.","location":"abstract s13","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":9759,"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":0.08,"states_limitation":0.84,"human_participants":0.26,"controlled_comparison":0.19,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.68,"mentions_code_or_data_availability":0.05},"evidential_basis":{"p":0.97,"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":0.97,"simulation_or_computation":0.03}},"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":1.0,"text":"We introduce HABIT-Bench to evaluate how well agent memory systems infer habits from weak evidence and apply them to later requests.","location":"abstract s4"}},"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":14,"sentences_in_abstract":14},"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"}}}