{"id":"https://openalex.org/W4416167273","doi":"https://doi.org/10.1109/iccv51701.2025.00086","title":"DocThinker: Explainable Multimodal Large Language Models with Rule-Based Reinforcement Learning for Document Understanding","display_name":"DocThinker: Explainable Multimodal Large Language Models with Rule-Based Reinforcement Learning for Document Understanding","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416167273","doi":"https://doi.org/10.1109/iccv51701.2025.00086"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.00086","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00086","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.08589","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5110949196","display_name":"Wenwen Yu","orcid":"https://orcid.org/0000-0003-1027-0090"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenwen Yu","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101866259","display_name":"Zhibo Yang","orcid":"https://orcid.org/0000-0002-9815-5013"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhibo Yang","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100389396","display_name":"Yuliang Liu","orcid":"https://orcid.org/0000-0002-3037-173X"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuliang Liu","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5039363991","display_name":"Xiang Bai","orcid":"https://orcid.org/0000-0002-3449-5940"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiang Bai","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.31613698,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"837","last_page":"847"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.44690001010894775,"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":0.44690001010894775,"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":0.19589999318122864,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.09350000321865082,"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/adaptability","display_name":"Adaptability","score":0.7488999962806702},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7184000015258789},{"id":"https://openalex.org/keywords/forgetting","display_name":"Forgetting","score":0.7077999711036682},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6983000040054321},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.4869000017642975},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.42829999327659607}],"concepts":[{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.7488999962806702},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7440999746322632},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7184000015258789},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.7077999711036682},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6983000040054321},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5828999876976013},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.4869000017642975},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.42829999327659607},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.362199991941452},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.36039999127388},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3276999890804291},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.30059999227523804},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.26489999890327454}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.00086","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00086","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2508.08589","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.08589","pdf_url":"https://arxiv.org/pdf/2508.08589","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2508.08589","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.08589","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2508.08589","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.08589","pdf_url":"https://arxiv.org/pdf/2508.08589","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5467188374","display_name":null,"funder_award_id":"62225603,62206104","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"Large":[1],"Language":[2],"Models":[3],"(MLLMs)":[4],"have":[5],"demonstrated":[6],"remarkable":[7],"capabilities":[8],"in":[9,28,166],"document":[10,37,168],"understanding.":[11,169],"However,":[12],"their":[13],"reasoning":[14,45,89,100,151],"processes":[15],"remain":[16],"largely":[17],"black-box,":[18],"making":[19],"it":[20],"difficult":[21],"to":[22],"ensure":[23],"reliability":[24],"and":[25,35,57,111,120,128,132,149,164],"trustworthiness,":[26],"especially":[27],"high-stakes":[29],"domains":[30],"such":[31],"as":[32,157],"legal,":[33],"financial,":[34],"medical":[36],"analysis.":[38],"Existing":[39],"methods":[40],"use":[41],"fixed":[42],"Chain-of-Thought":[43],"(CoT)":[44],"with":[46],"supervised":[47],"fine-tuning":[48],"(SFT)":[49],"but":[50],"suffer":[51],"from":[52],"catastrophic":[53,126],"forgetting,":[54],"poor":[55],"adaptability,":[56],"limited":[58],"generalization":[59,144],"across":[60],"domain":[61],"tasks.":[62],"In":[63],"this":[64],"paper,":[65],"we":[66],"propose":[67],"DocThinker,":[68],"a":[69,158],"rule-based":[70,118],"Reinforcement":[71],"Learning":[72],"(RL)":[73],"framework":[74],"for":[75,161],"dynamic":[76],"inference-time":[77],"reasoning.":[78],"Instead":[79],"of":[80,105],"relying":[81],"on":[82,136],"static":[83],"CoT":[84],"templates,":[85],"DocThinker":[86,141],"autonomously":[87],"refines":[88],"strategies":[90],"via":[91],"policy":[92],"learning,":[93],"generating":[94],"explainable":[95,148],"intermediate":[96],"results,":[97],"including":[98],"structured":[99],"processes,":[101],"rephrased":[102],"questions,":[103],"regions":[104],"interest":[106],"(RoI)":[107],"supporting":[108],"the":[109,112],"answer,":[110],"final":[113],"answer.":[114],"By":[115],"integrating":[116],"multi-objective":[117],"rewards":[119],"KL-constrained":[121],"optimization,":[122],"our":[123],"method":[124],"mitigates":[125],"forgetting":[127],"enhances":[129],"both":[130],"adaptability":[131,165],"transparency.":[133],"Extensive":[134],"experiments":[135],"multiple":[137],"benchmarks":[138],"demonstrate":[139],"that":[140],"significantly":[142],"improves":[143],"while":[145],"producing":[146],"more":[147],"human-understandable":[150],"steps.":[152],"Our":[153],"findings":[154],"highlight":[155],"RL":[156],"powerful":[159],"alternative":[160],"enhancing":[162],"explainability":[163],"MLLM-based":[167],"Code":[170],"will":[171],"be":[172],"available":[173],"at":[174],"https://github.com/wenwenyu/DocThinker.":[175]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}