{"paper":{"title":"MemDev-Bench: Does Memory Help Code Agents?Benchmarking Memory Systems over Real Repository Histories","arxiv_id":"2610.27524","primary_cat":"cs.AI","submitted_at":"2026-10-20T02:39:26+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":"Code agents increasingly work on evolving real repositories, and memory systems aim to help them reuse earlier experience.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.04,"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":0.96,"id":"s1","own":false,"role":"gap","text":"However, such experience is diverse and evolving, so it remains unclear how memory systems organize it and thereby affect code agents.","hedge":{"p":0.7,"label":"asserted"},"own_p":0.05,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.96,"method":0.0,"result":0.0,"background":0.04,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.99,"id":"s2","own":false,"role":"gap","text":"Existing benchmarks cannot answer this reliably because they lack either temporally valid histories or control over prior-task relevance.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.09,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.99,"method":0.0,"result":0.0,"background":0.01,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"contribution","text":"In this paper, we present MemDev-Bench to evaluate how memory systems affect code agents over diverse repository histories.","hedge":{"p":0.97,"label":"asserted"},"own_p":0.93,"main_claim_p":0.04,"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":4,"p":0.97,"id":"s4","own":true,"role":"method","text":"It pairs each task with relevance-annotated earlier tasks from the same repository to build histories with controlled relevance.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.97,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.03,"interpretation":0.0}},{"i":5,"p":0.96,"id":"s5","own":true,"role":"method","text":"We also build an automated pipeline that reliably makes these earlier tasks executable and verifiable, producing 4,869 historical tasks for 555 base tasks.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.98,"main_claim_p":0.02,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.96,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.04,"interpretation":0.0}},{"i":6,"p":0.62,"id":"s6","own":true,"role":"result","text":"Based on MemDev-Bench, our systematic evaluation of memory systems across diverse development scenarios yields three main findings.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.94,"main_claim_p":0.09,"quantitative":true,"quantitative_p":0.7,"role_probabilities":{"gap":0.0,"method":0.01,"result":0.62,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.36,"interpretation":0.01}},{"i":7,"p":1.0,"id":"s7","own":true,"role":"result","text":"(1) Memory mainly improves fine-grained localization and editing, increasing functional correctness.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.23,"quantitative":false,"quantitative_p":0.13,"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":8,"p":1.0,"id":"s8","own":true,"role":"result","text":"(2) Memory also causes failures on otherwise solvable tasks even with related history, and unrelated history tends to increase such failures.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.17,"quantitative":false,"quantitative_p":0.16,"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":9,"p":0.91,"id":"s9","own":true,"role":"interpretation","text":"These failures mainly stem from agents reusing memory without reassessing its applicability.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.87,"main_claim_p":0.11,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.09,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.91}},{"i":10,"p":1.0,"id":"s10","own":true,"role":"result","text":"(3) Compared with using memory systems, directly providing agents with raw historical trajectories substantially reduces their steps.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.19,"quantitative":true,"quantitative_p":0.89,"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":11,"p":0.95,"id":"s11","own":true,"role":"interpretation","text":"Further analysis suggests that agents save these steps by reusing process information that compressed memories tend to omit.","hedge":{"p":0.97,"label":"hedged_possibility"},"own_p":0.88,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.14,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.05,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.95}},{"i":12,"p":0.91,"id":"s12","own":true,"role":"future_work","text":"These results suggest that future memory systems should preserve process information and reassess memory applicability against current task context.","hedge":{"p":0.99,"label":"hedged_possibility"},"own_p":0.6,"main_claim_p":0.72,"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":0.91,"contribution":0.0,"interpretation":0.09}}]},"compiled":{"main_claim":{"p":0.97,"text":"These results suggest that future memory systems should preserve process information and reassess memory applicability against current task context.","location":"abstract s12","hedge_strength":{"p":0.99,"label":"hedged possibility"}},"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":9310,"cost_usd_reported":0.0},"compiled_at":"2026-10-21","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.12,"states_limitation":0.24,"human_participants":0.05,"controlled_comparison":0.89,"randomized_assignment":0.02,"states_quantitative_result":0.96,"compares_with_prior_results":0.9,"mentions_code_or_data_availability":0.06},"evidential_basis":{"p":1.0,"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":1.0,"simulation_or_computation":0.0}},"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":0.97,"text":"In this paper, we present MemDev-Bench to evaluate how memory systems affect code agents over diverse repository histories.","location":"abstract s3"}},"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":13,"sentences_in_abstract":13},"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"}}}