{"id":"https://openalex.org/W4414578351","doi":"https://doi.org/10.1109/iccv51701.2025.01752","title":"GTR: Guided Thought Reinforcement Prevents Thought Collapse in RL-Based VLM Agent Training","display_name":"GTR: Guided Thought Reinforcement Prevents Thought Collapse in RL-Based VLM Agent Training","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4414578351","doi":"https://doi.org/10.1109/iccv51701.2025.01752"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.01752","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01752","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/2503.08525","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022293237","display_name":"Wei Tu","orcid":"https://orcid.org/0000-0002-0255-4037"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tong Wei","raw_affiliation_strings":["Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054969834","display_name":"Yijun Yang","orcid":"https://orcid.org/0000-0002-9243-6228"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yijun Yang","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076090670","display_name":"Junliang Xing","orcid":"https://orcid.org/0000-0001-6801-0510"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junliang Xing","raw_affiliation_strings":["Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057896400","display_name":"Yuanchun Shi","orcid":"https://orcid.org/0000-0003-2273-6927"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuanchun Shi","raw_affiliation_strings":["Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020361550","display_name":"Zongqing Lu","orcid":"https://orcid.org/0000-0002-1191-9069"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zongqing Lu","raw_affiliation_strings":["Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5073681676","display_name":"Deheng Ye","orcid":"https://orcid.org/0000-0002-1754-1837"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Deheng Ye","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":"18855","last_page":"18865"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10429","display_name":"EEG and Brain-Computer Interfaces","score":0.9253000020980835,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T10429","display_name":"EEG and Brain-Computer Interfaces","score":0.9253000020980835,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7041000127792358},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.6553000211715698},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5838000178337097},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5626000165939331},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5356000065803528},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.46470001339912415},{"id":"https://openalex.org/keywords/outcome","display_name":"Outcome (game theory)","score":0.4142000079154968},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.40369999408721924}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7041000127792358},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.6553000211715698},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5838000178337097},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5626000165939331},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5356000065803528},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5127999782562256},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5123000144958496},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.46470001339912415},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.4296000003814697},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.4142000079154968},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.40369999408721924},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3991999924182892},{"id":"https://openalex.org/C132758656","wikidata":"https://www.wikidata.org/wiki/Q5307365","display_name":"Dreyfus model of skill acquisition","level":2,"score":0.34769999980926514},{"id":"https://openalex.org/C50335755","wikidata":"https://www.wikidata.org/wiki/Q483247","display_name":"Phenomenon","level":2,"score":0.3312999904155731},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3292999863624573},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.32739999890327454},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.3255000114440918},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.2896000146865845},{"id":"https://openalex.org/C176777502","wikidata":"https://www.wikidata.org/wiki/Q4774623","display_name":"Anticipation (artificial intelligence)","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2802000045776367},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.25760000944137573},{"id":"https://openalex.org/C2982912361","wikidata":"https://www.wikidata.org/wiki/Q1851867","display_name":"Mental model","level":2,"score":0.25369998812675476}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.01752","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01752","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:2503.08525","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.08525","pdf_url":"https://arxiv.org/pdf/2503.08525","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.2503.08525","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.08525","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":"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":"pmh:oai:arXiv.org:2503.08525","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.08525","pdf_url":"https://arxiv.org/pdf/2503.08525","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":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"with":[2,189],"verifiable":[3],"outcome":[4],"rewards":[5,63],"(RLVR)":[6],"has":[7],"effectively":[8],"scaled":[9],"up":[10],"chain-of-thought":[11],"(CoT)":[12],"reasoning":[13,31,75,132,147],"in":[14,22,32,76,94,107],"large":[15],"language":[16],"models":[17,188],"(LLMs).":[18],"Yet,":[19],"its":[20],"efficacy":[21],"training":[23],"vision-language":[24],"model":[25,173,192],"(VLM)":[26],"agents":[27],"for":[28,154],"goal-directed":[29],"action":[30,68,149],"visual":[33,176],"environments":[34],"is":[35],"less":[36],"established.":[37],"This":[38,137],"work":[39],"investigates":[40],"this":[41],"problem":[42],"through":[43],"extensive":[44],"experiments":[45,160],"on":[46,67],"complex":[47],"card":[48],"games,":[49],"such":[50],"as":[51],"24":[52],"points,":[53],"and":[54,99,102,121,128,139,148,168],"embodied":[55],"tasks":[56],"from":[57],"ALFWorld.":[58],"We":[59],"find":[60],"that":[61,126,162],"when":[62],"are":[64],"based":[65],"solely":[66],"outcomes,":[69],"RL":[70,135],"fails":[71],"to":[72,80,186],"incentivize":[73],"CoT":[74],"VLMs,":[77],"instead":[78],"leading":[79],"a":[81,89],"phenomenon":[82],"we":[83,114],"termed":[84],"thought":[85,112],"collapse,":[86,113],"characterized":[87],"by":[88],"rapid":[90],"loss":[91],"of":[92,118,170],"diversity":[93],"the":[95,116,130,152,166,171],"agent's":[96,131],"thoughts,":[97],"state-irrelevant":[98],"incomplete":[100],"reasoning,":[101],"subsequent":[103],"invalid":[104],"actions,":[105],"resulting":[106],"negative":[108],"rewards.":[109],"To":[110],"counteract":[111],"highlight":[115],"necessity":[117],"process":[119],"guidance":[120],"propose":[122],"an":[123],"automated":[124],"corrector":[125],"evaluates":[127],"refines":[129],"at":[133],"each":[134],"step.":[136],"simple":[138],"scalable":[140],"GTR":[141,163],"(Guided":[142],"Thought":[143],"Reinforcement)":[144],"framework":[145],"trains":[146],"simultaneously":[150],"without":[151],"need":[153],"dense,":[155],"per-step":[156],"human":[157],"labeling.":[158],"Our":[159],"demonstrate":[161],"significantly":[164],"enhances":[165],"performance":[167],"generalization":[169],"LLaVA-7b":[172],"across":[174],"various":[175],"environments,":[177],"achieving":[178],"3-5":[179],"times":[180],"higher":[181],"task":[182],"success":[183],"rates":[184],"compared":[185],"SoTA":[187],"notably":[190],"smaller":[191],"sizes.":[193]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-10-10T00:00:00"}