{"id":"https://openalex.org/W7225802448","doi":"https://doi.org/10.48550/arxiv.2611.13907","title":"EVAThink: Evidence Sufficiency-Guided Training for Adaptive Thinking in Video Understanding","display_name":"EVAThink: Evidence Sufficiency-Guided Training for Adaptive Thinking in Video Understanding","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225802448","doi":"https://doi.org/10.48550/arxiv.2611.13907"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2611.13907","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.13907","pdf_url":"https://arxiv.org/pdf/2611.13907","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,"listed_in":[],"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"},"parent_works":[],"lineage":["https://openalex.org/W7225802448"],"child_works_count":0,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2611.13907","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069583018","display_name":"Changti Wu","orcid":"https://orcid.org/0009-0009-9448-6657","observed_orcids":["https://orcid.org/0009-0009-9448-6657"]},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Changti","raw_affiliation_strings":["East China Normal University","Zhongguancun Academy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"East China Normal University","institution_ids":["https://openalex.org/I66867065"]},{"raw_affiliation_string":"Zhongguancun Academy","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161472044","display_name":"Ruizhe Fan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan, Ruizhe","raw_affiliation_strings":["Huawei Technologies Co. , Ltd"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Co. , Ltd","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161423338","display_name":"Bin Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Bin","raw_affiliation_strings":["Huawei Technologies Co. , Ltd"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Co. , Ltd","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161436098","display_name":"Lei Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Lei","raw_affiliation_strings":["East China Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"East China Normal University","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5161480863","display_name":"Yuxiang Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yuxiang","raw_affiliation_strings":["Huawei Technologies Co. , Ltd"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Co. , Ltd","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.0,"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"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0,"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/video-understanding","display_name":"video understanding","score":0.9779999852180481},{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.8889999985694885},{"id":"https://openalex.org/keywords/frame-sampling","display_name":"frame sampling","score":0.546999990940094},{"id":"https://openalex.org/keywords/sufficiency-of-evidence","display_name":"sufficiency of evidence","score":0.9399999976158142},{"id":"https://openalex.org/keywords/adaptive-thinking","display_name":"adaptive thinking","score":0.5899999737739563},{"id":"https://openalex.org/keywords/mode-collapse","display_name":"mode collapse","score":0.7590000033378601}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7160999774932861},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.5723000168800354},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49239999055862427},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.44749999046325684},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4431000053882599},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44279998540878296},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.4083999991416931},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3853999972343445},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.3257000148296356},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.32359999418258667},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.28679999709129333},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.27410000562667847},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2611.13907","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.13907","pdf_url":"https://arxiv.org/pdf/2611.13907","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,"listed_in":[],"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.2611.13907","doi":"https://doi.org/10.48550/arxiv.2611.13907","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.13907","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2611.13907","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.13907","pdf_url":"https://arxiv.org/pdf/2611.13907","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,"listed_in":[],"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225802448.pdf","grobid_xml":"https://content.openalex.org/works/W7225802448.grobid-xml"},"referenced_works_count":17,"referenced_works":["https://openalex.org/W4391631327","https://openalex.org/W4414281281","https://openalex.org/W4414898865","https://openalex.org/W4415320074","https://openalex.org/W4416780398","https://openalex.org/W6948109617","https://openalex.org/W6948316781","https://openalex.org/W7083815686","https://openalex.org/W7101754354","https://openalex.org/W7120112120","https://openalex.org/W7123566719","https://openalex.org/W7128807805","https://openalex.org/W7161580828","https://openalex.org/W7166493911","https://openalex.org/W7172402630","https://openalex.org/W7202222885","https://openalex.org/W7204552031"],"related_works":[],"abstract_inverted_index":{"Under":[0],"a":[1],"fixed":[2],"frame":[3,24,77],"budget,":[4],"video":[5,47,65,143],"understanding":[6,48,144],"with":[7,113],"vision-language":[8],"models":[9],"(VLMs)":[10],"depends":[11],"on":[12,130,156],"whether":[13],"the":[14,21,39,69,139],"sampled":[15],"frames":[16],"provide":[17],"sufficient":[18],"evidence":[19,25,50],"for":[20,53,99],"question.":[22],"Sufficient":[23],"supports":[26],"direct":[27],"answering,":[28],"whereas":[29],"less":[30],"informative":[31],"observations":[32],"may":[33],"benefit":[34,93],"from":[35,94],"reasoning":[36],"that":[37,67],"connects":[38],"available":[40,164],"cues.":[41],"We":[42],"introduce":[43],"EVAThink,":[44],"which":[45],"enhances":[46],"through":[49],"sufficiency-guided":[51],"training":[52,112],"adaptive":[54],"thinking.":[55],"First,":[56],"Recoverable":[57],"Evidence":[58],"Gap":[59],"Mining":[60],"constructs":[61],"evidence-dense":[62],"and":[63,72,117,149,161],"evidence-sparse":[64],"views":[66,84,91],"share":[68],"same":[70],"question":[71],"answer":[73],"under":[74,146],"an":[75],"identical":[76],"budget.":[78],"It":[79],"retains":[80],"pairs":[81],"where":[82],"sparse":[83],"yield":[85],"lower":[86],"direct-answer":[87],"accuracy":[88],"than":[89],"dense":[90],"but":[92],"explicit":[95],"reasoning,":[96],"providing":[97],"supervision":[98],"response-mode":[100],"selection.":[101],"Second,":[102],"Evidence-Guided":[103],"Two-Stage":[104],"Reinforcement":[105],"Learning":[106],"uses":[107],"stage-specific":[108],"rewards":[109],"to":[110],"guide":[111],"evidence-based":[114],"mode":[115,127],"preferences":[116],"then":[118],"refine":[119],"response":[120],"efficiency,":[121],"while":[122,152],"first-token":[123],"advantage":[124],"regularization":[125],"mitigates":[126],"collapse.":[128],"Trained":[129],"only":[131],"6.7K":[132],"curated":[133],"samples,":[134],"EVAThink":[135],"consistently":[136],"improves":[137],"upon":[138],"base":[140],"model":[141],"across":[142],"benchmarks":[145],"both":[147],"32-":[148],"64-frame":[150],"budgets,":[151],"outperforming":[153],"strong":[154],"baselines":[155],"average.":[157],"The":[158],"code,":[159],"model,":[160],"datasets":[162],"are":[163],"at":[165],"\\href{https://github.com/ChangtiWu/EVAThink}{EVAThink}.":[166]},"counts_by_year":[],"updated_date":"2026-11-14T11:04:58.165337","created_date":"2026-11-12T00:00:00"}