{"id":"https://openalex.org/W7148462007","doi":"https://doi.org/10.48550/arxiv.2604.00013","title":"C2F-Thinker: Coarse-to-Fine Reasoning with Hint-Guided Reinforcement Learning for Multimodal Sentiment Analysis","display_name":"C2F-Thinker: Coarse-to-Fine Reasoning with Hint-Guided Reinforcement Learning for Multimodal Sentiment Analysis","publication_year":2026,"publication_date":"2026-03-10","ids":{"openalex":"https://openalex.org/W7148462007","doi":"https://doi.org/10.48550/arxiv.2604.00013"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.00013","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00013","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.00013","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5132817659","display_name":"Miaosen Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Luo, Miaosen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102522900","display_name":"Zhenhao Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Zhenhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132825615","display_name":"Jieshen Long","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Long, Jieshen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067039090","display_name":"Jinghu Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Jinghu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083557186","display_name":"Yichu Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yichu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5010270301","display_name":"Sijie Mai","orcid":"https://orcid.org/0000-0001-9763-375X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mai, Sijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5132817659"],"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":0.8964999914169312,"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":0.8964999914169312,"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/T10667","display_name":"Emotion and Mood Recognition","score":0.028200000524520874,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.01489999983459711,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.8486999869346619},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.737500011920929},{"id":"https://openalex.org/keywords/sentiment-analysis","display_name":"Sentiment analysis","score":0.6283000111579895},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5497999787330627},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.424699991941452},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.38199999928474426},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.38190001249313354},{"id":"https://openalex.org/keywords/macro","display_name":"Macro","score":0.37869998812675476}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8486999869346619},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7741000056266785},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.737500011920929},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6761000156402588},{"id":"https://openalex.org/C66402592","wikidata":"https://www.wikidata.org/wiki/Q2271421","display_name":"Sentiment analysis","level":2,"score":0.6283000111579895},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5543000102043152},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5497999787330627},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.424699991941452},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.38199999928474426},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.38190001249313354},{"id":"https://openalex.org/C166955791","wikidata":"https://www.wikidata.org/wiki/Q629579","display_name":"Macro","level":2,"score":0.37869998812675476},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3569999933242798},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.27810001373291016},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2759999930858612},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.27309998869895935},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2709999978542328},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2669000029563904},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.00013","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00013","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.00013","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00013","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":"article"},"sustainable_development_goals":[{"score":0.5721246004104614,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"sentiment":[1,204,223],"analysis":[2,224],"aims":[3],"to":[4,187],"integrate":[5],"textual,":[6],"acoustic,":[7],"and":[8,48,60,115,167,182,221],"visual":[9],"information":[10],"for":[11,226],"deep":[12],"emotional":[13,126],"understanding.":[14],"Despite":[15],"the":[16,49,88,120,130,152,155,169],"progress":[17],"of":[18,52,107,171],"multimodal":[19],"large":[20],"language":[21],"models":[22],"(MLLMs)":[23],"via":[24],"supervised":[25,94],"fine-tuning,":[26],"their":[27],"\"black-box\"":[28],"nature":[29],"hinders":[30],"interpretability.":[31,193],"While":[32],"Chain-of-Thought":[33],"(CoT)":[34],"reasoning":[35,77,127,161],"offers":[36],"a":[37,71,82,102,124,135,175],"potential":[38,217],"remedy,":[39],"it":[40],"is":[41,157,185],"constrained":[42],"by":[43],"high":[44],"manual":[45],"annotation":[46],"costs":[47],"inherent":[50],"challenges":[51],"reinforcement":[53],"learning":[54],"(RL),":[55],"such":[56],"as":[57,149],"reward":[58,177],"sparsity":[59],"low":[61],"exploration":[62],"efficiency":[63],"on":[64,202],"hard":[65,172],"samples.":[66,173],"This":[67,118,214],"paper":[68],"presents":[69],"C2F-Thinker,":[70],"framework":[72],"that":[73,197],"harmonizes":[74],"coarse-to-fine":[75],"structured":[76,125],"with":[78,123],"hint-guided":[79,136],"RL":[80],"through":[81],"two-stage":[83],"progressive":[84],"training":[85],"pipeline.":[86],"In":[87,129],"first":[89],"stage,":[90,132],"we":[91,133],"conduct":[92],"cold-start":[93],"fine-tuning":[95],"using":[96],"high-quality":[97],"CoT":[98],"data":[99],"distilled":[100],"from":[101],"larger":[103],"teacher":[104],"model,":[105],"consisting":[106],"three":[108],"distinct":[109],"phases:":[110],"polarity":[111,147],"judgment,":[112],"intermediate":[113],"analysis,":[114],"fine-grained":[116,203],"scoring.":[117],"equips":[119],"base":[121],"model":[122,156],"paradigm.":[128],"second":[131],"introduce":[134],"Group":[137],"Relative":[138],"Policy":[139],"Optimization":[140],"(GRPO)":[141],"algorithm.":[142],"By":[143],"injecting":[144],"correct":[145],"initial":[146],"predictions":[148],"hints":[150],"during":[151],"sampling":[153],"process,":[154],"guided":[158],"toward":[159],"accurate":[160],"paths,":[162],"effectively":[163],"mitigating":[164],"cascading":[165],"errors":[166],"enhancing":[168],"utilization":[170],"Furthermore,":[174],"multi-faceted":[176],"function":[178],"incorporating":[179],"classification,":[180],"regression,":[181],"formatting":[183],"constraints":[184],"designed":[186],"refine":[188],"prediction":[189],"accuracy":[190],"while":[191,207],"preserving":[192],"Experimental":[194],"results":[195],"demonstrate":[196],"C2F-Thinker":[198],"achieves":[199],"competitive":[200],"performance":[201],"regression":[205],"tasks":[206],"significantly":[208],"outperforming":[209],"baselines":[210],"in":[211,218],"cross-domain":[212],"generalization.":[213],"highlights":[215],"its":[216],"building":[219],"trustworthy":[220],"robust":[222],"systems":[225],"real-world":[227],"applications.":[228]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2026-04-03T00:00:00"}