{"id":"https://openalex.org/W7154566900","doi":"https://doi.org/10.48550/arxiv.2604.13398","title":"From Prediction to Justification: Aligning Sentiment Reasoning with Human Rationale via Reinforcement Learning","display_name":"From Prediction to Justification: Aligning Sentiment Reasoning with Human Rationale via Reinforcement Learning","publication_year":2026,"publication_date":"2026-04-15","ids":{"openalex":"https://openalex.org/W7154566900","doi":"https://doi.org/10.48550/arxiv.2604.13398"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.13398","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.13398","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2604.13398","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100624693","display_name":"Shihao Zhang","orcid":"https://orcid.org/0000-0003-0984-9978"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133769112","display_name":"Ziwei Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ziwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133785863","display_name":"Jie Zhou","orcid":"https://orcid.org/0009-0009-9232-9631"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133772002","display_name":"Yulan Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yulan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133750240","display_name":"Qin Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Qin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133748195","display_name":"Zhikai Lei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei, Zhikai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133804614","display_name":"Liyang Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Liyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100668340","display_name":"Liang Dou","orcid":"https://orcid.org/0000-0003-3044-3841"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dou, Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133809983","display_name":"Liang He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Liang","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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.9160000085830688,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.9160000085830688,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.0203000009059906,"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/T10028","display_name":"Topic Modeling","score":0.011800000444054604,"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/interpretability","display_name":"Interpretability","score":0.8598999977111816},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6389999985694885},{"id":"https://openalex.org/keywords/categorization","display_name":"Categorization","score":0.6050999760627747},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.5758000016212463},{"id":"https://openalex.org/keywords/sentiment-analysis","display_name":"Sentiment analysis","score":0.5162000060081482},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5005999803543091},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.4544999897480011},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"Commonsense reasoning","score":0.43459999561309814}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8598999977111816},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7573999762535095},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6873999834060669},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6389999985694885},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.6050999760627747},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5758000016212463},{"id":"https://openalex.org/C66402592","wikidata":"https://www.wikidata.org/wiki/Q2271421","display_name":"Sentiment analysis","level":2,"score":0.5162000060081482},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5005999803543091},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.4544999897480011},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4408999979496002},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.43459999561309814},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.42980000376701355},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.40130001306533813},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3765999972820282},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.36970001459121704},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.31619998812675476},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.29899999499320984},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.29159998893737793},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2687999904155731},{"id":"https://openalex.org/C48164120","wikidata":"https://www.wikidata.org/wiki/Q4491893","display_name":"Concept learning","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C149271511","wikidata":"https://www.wikidata.org/wiki/Q1417149","display_name":"Rule-based system","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.13398","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.13398","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.13398","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.13398","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.48386383056640625}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"While":[0],"Aspect-based":[1],"Sentiment":[2],"Analysis":[3],"(ABSA)":[4],"systems":[5],"have":[6],"achieved":[7],"high":[8],"accuracy":[9],"in":[10,157],"identifying":[11],"sentiment":[12,83,158],"polarities,":[13],"they":[14,36],"often":[15],"operate":[16],"as":[17],"\"black":[18],"boxes,\"":[19],"lacking":[20],"the":[21,71,74,99,104,126],"explicit":[22,145],"reasoning":[23,101,129,146],"capabilities":[24],"characteristic":[25],"of":[26],"human":[27],"affective":[28],"cognition.":[29],"Humans":[30],"do":[31],"not":[32,148],"merely":[33],"categorize":[34],"sentiment;":[35],"construct":[37],"causal":[38],"explanations":[39],"for":[40],"their":[41],"judgments.":[42],"To":[43],"bridge":[44],"this":[45,58,144],"gap,":[46],"we":[47,113],"propose":[48],"ABSA-R1,":[49],"a":[50,87,115],"large":[51],"language":[52,78],"model":[53],"framework":[54],"designed":[55],"to":[56,69,164],"mimic":[57],"``reason-before-predict\"":[59],"cognitive":[60],"process.":[61],"By":[62],"leveraging":[63],"reinforcement":[64],"learning":[65],"(RL),":[66],"ABSA-R1":[67],"learns":[68],"articulate":[70],"why":[72],"behind":[73],"what,":[75],"generating":[76],"natural":[77],"justifications":[79],"that":[80,95,120,140],"ground":[81],"its":[82],"predictions.":[84],"We":[85],"introduce":[86],"Cognition-Aligned":[88],"Reward":[89],"Model":[90],"(formerly":[91],"sentiment-aware":[92],"reward":[93],"model)":[94],"enforces":[96],"consistency":[97],"between":[98],"generated":[100],"path":[102],"and":[103,160],"final":[105],"emotional":[106],"label.":[107],"Furthermore,":[108],"inspired":[109],"by":[110],"metacognitive":[111],"monitoring,":[112],"implement":[114],"performance-driven":[116],"rejection":[117],"sampling":[118],"strategy":[119],"selectively":[121],"targets":[122],"hard":[123],"cases":[124],"where":[125],"model's":[127],"internal":[128],"is":[130],"uncertain":[131],"or":[132],"inconsistent.":[133],"Experimental":[134],"results":[135],"on":[136],"four":[137],"benchmarks":[138],"demonstrate":[139],"equipping":[141],"models":[142],"with":[143],"capability":[147],"only":[149],"enhances":[150],"interpretability":[151],"but":[152],"also":[153],"yields":[154],"superior":[155],"performance":[156],"classification":[159],"triplet":[161],"extraction":[162],"compared":[163],"non-reasoning":[165],"baselines.":[166]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-17T00:00:00"}