{"id":"https://openalex.org/W7154270697","doi":"https://doi.org/10.48550/arxiv.2604.11297","title":"The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping","display_name":"The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping","publication_year":2026,"publication_date":"2026-04-13","ids":{"openalex":"https://openalex.org/W7154270697","doi":"https://doi.org/10.48550/arxiv.2604.11297"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.11297","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.11297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.11297","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133621539","display_name":"Yang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133624178","display_name":"Enxi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Enxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133581119","display_name":"Yufei Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Yufei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133573999","display_name":"Weixin Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Weixin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080515342","display_name":"Bo Wang","orcid":"https://orcid.org/0000-0002-0466-6323"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133576663","display_name":"Zhiyuan Zeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Zhiyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133587408","display_name":"Yikai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yikai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133601544","display_name":"Yining Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Yining","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133556704","display_name":"Xipeng Qiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiu, Xipeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.33410000801086426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.33410000801086426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.19030000269412994,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13629","display_name":"Text Readability and Simplification","score":0.04659999907016754,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/randomness","display_name":"Randomness","score":0.657800018787384},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6008999943733215},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.51419997215271},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.49950000643730164},{"id":"https://openalex.org/keywords/diversity","display_name":"Diversity (politics)","score":0.4023999869823456},{"id":"https://openalex.org/keywords/mean-squared-prediction-error","display_name":"Mean squared prediction error","score":0.34360000491142273}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7299000024795532},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.657800018787384},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6008999943733215},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.51419997215271},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.49950000643730164},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4812999963760376},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47600001096725464},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.4023999869823456},{"id":"https://openalex.org/C167085575","wikidata":"https://www.wikidata.org/wiki/Q6803654","display_name":"Mean squared prediction error","level":2,"score":0.34360000491142273},{"id":"https://openalex.org/C73602740","wikidata":"https://www.wikidata.org/wiki/Q7795822","display_name":"Thompson sampling","level":3,"score":0.337799996137619},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.30809998512268066},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.30309998989105225},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.2890999913215637},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.11297","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.11297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.11297","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.11297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Despite":[0],"the":[1,19,32],"success":[2],"of":[3,72,120],"reinforcement":[4],"learning":[5],"for":[6],"large":[7],"language":[8],"models,":[9,109],"a":[10,48],"common":[11],"failure":[12,41],"mode":[13],"is":[14],"reduced":[15],"sampling":[16],"diversity,":[17],"where":[18],"policy":[20],"repeatedly":[21],"generates":[22],"similar":[23],"erroneous":[24],"behaviors.":[25],"Classical":[26],"entropy":[27],"regularization":[28],"encourages":[29],"randomness":[30],"under":[31],"current":[33],"policy,":[34],"but":[35],"does":[36],"not":[37],"explicitly":[38],"discourage":[39],"recurrent":[40],"patterns":[42],"across":[43],"rollouts.":[44],"We":[45],"propose":[46],"MEDS,":[47],"Memory-Enhanced":[49],"Dynamic":[50],"reward":[51,60],"Shaping":[52],"framework":[53],"that":[54,141],"incorporates":[55],"historical":[56],"behavioral":[57,144],"signals":[58],"into":[59],"design.":[61],"By":[62],"storing":[63],"and":[64,75,106,126,136],"leveraging":[65],"intermediate":[66],"model":[67],"representations,":[68],"we":[69],"capture":[70],"features":[71],"past":[73],"rollouts":[74],"use":[76],"density-based":[77],"clustering":[78],"to":[79,87,122],"identify":[80],"frequently":[81],"recurring":[82],"error":[83,90],"patterns.":[84],"Rollouts":[85],"assigned":[86],"more":[88,94],"prevalent":[89],"clusters":[91],"are":[92],"penalized":[93],"heavily,":[95],"encouraging":[96],"broader":[97],"exploration":[98],"while":[99],"reducing":[100],"repeated":[101],"mistakes.":[102],"Across":[103],"five":[104],"datasets":[105],"three":[107],"base":[108],"MEDS":[110,142],"consistently":[111],"improves":[112],"average":[113],"performance":[114],"over":[115],"existing":[116],"baselines,":[117],"achieving":[118],"gains":[119],"up":[121],"4.13":[123],"pass@1":[124],"points":[125],"4.37":[127],"pass@128":[128],"points.":[129],"Additional":[130],"analyses":[131],"using":[132],"both":[133],"LLM-based":[134],"annotations":[135],"quantitative":[137],"diversity":[138,145],"metrics":[139],"show":[140],"increases":[143],"during":[146],"sampling.":[147]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-15T00:00:00"}