{"paper":{"title":"HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agents","arxiv_id":"2606.16285","primary_cat":"cs.CL","submitted_at":"2026-06-15T06:49:18+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":0.87,"id":"s0","own":false,"role":"background","text":"Long-horizon agents rely on memory mechanisms to compress interaction history, but optimizing memory writing faces a distinct credit assignment challenge: a memory update may be rewarded or penalized due to downstream tool failures, noisy observations, or reasoning errors rather than its own contribution.","hedge":{"p":0.74,"label":"asserted"},"own_p":0.09,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.13,"method":0.0,"result":0.0,"background":0.87,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.99,"id":"s1","own":true,"role":"contribution","text":"We propose HiMPO, a Hindsight-Informed Memory Policy Optimization framework for assigning less-entangled credit to memory-writing actions in long-horizon agents.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.83,"main_claim_p":0.17,"quantitative":false,"quantitative_p":0.04,"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":2,"p":0.99,"id":"s2","own":true,"role":"method","text":"HiMPO first estimates the local utility of a memory update by comparing the task-relevant information recoverable from the previous and updated memories under the same pre-write state.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.82,"main_claim_p":0.04,"quantitative":true,"quantitative_p":0.52,"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":3,"p":0.99,"id":"s3","own":true,"role":"method","text":"It then uses hindsight relevance as a bounded retrospective filter that attenuates memory credit when local utility is not supported by the target outcome.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.83,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.42,"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":4,"p":0.99,"id":"s4","own":true,"role":"method","text":"The resulting memory-specific advantage is applied only to memory tokens, while trajectory-level rewards optimize the rest of the agent's behavior.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.79,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.09,"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":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"Across judge-based open-domain tasks and objective compressive-memory QA, HiMPO improves over strong memory-based and RL-based baselines while preserving compressed-context efficiency.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.56,"quantitative":true,"quantitative_p":0.81,"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":"Controlled interventions and live replay studies further show that HiMPO reduces blame leakage from tool-induced errors, assigns memory credit that aligns with the functional impact of memory writes, and remains robust to noisy training targets.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.25,"quantitative":false,"quantitative_p":0.1,"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}}]},"compiled":{"main_claim":{"p":0.94,"text":"Across judge-based open-domain tasks and objective compressive-memory QA, HiMPO improves over strong memory-based and RL-based baselines while preserving compressed-context efficiency.","location":"abstract s5","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":5718,"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":0.06,"states_limitation":0.11,"human_participants":0.1,"controlled_comparison":0.92,"randomized_assignment":0.02,"states_quantitative_result":0.05,"compares_with_prior_results":0.97,"mentions_code_or_data_availability":0.03},"evidential_basis":{"p":0.93,"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":0.93,"simulation_or_computation":0.07}},"contribution_type":{"p":1.0,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":1.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":1.0,"text":"We propose HiMPO, a Hindsight-Informed Memory Policy Optimization framework for assigning less-entangled credit to memory-writing actions in long-horizon agents.","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":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"}}}