{"id":"https://openalex.org/W7148963043","doi":"https://doi.org/10.48550/arxiv.2604.02040","title":"Efficient Reasoning via Thought Compression for Language Segmentation","display_name":"Efficient Reasoning via Thought Compression for Language Segmentation","publication_year":2026,"publication_date":"2026-04-02","ids":{"openalex":"https://openalex.org/W7148963043","doi":"https://doi.org/10.48550/arxiv.2604.02040"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.02040","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02040","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.02040","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5132900033","display_name":"Qing Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Qing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100409612","display_name":"Shiyu Zhang","orcid":"https://orcid.org/0000-0002-0302-824X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shiyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029861746","display_name":"Yuyu Jia","orcid":"https://orcid.org/0009-0002-2303-8995"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Yuyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132848505","display_name":"Junyu Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Junyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127789117","display_name":"Weiping Ni","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ni, Weiping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132870482","display_name":"Junzheng Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Junzheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5132842220","display_name":"Qi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Qi","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8328999876976013,"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":0.8328999876976013,"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/T10028","display_name":"Topic Modeling","score":0.04780000075697899,"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":0.03539999946951866,"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/inference","display_name":"Inference","score":0.61080002784729},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5303999781608582},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.46540001034736633},{"id":"https://openalex.org/keywords/imperfect","display_name":"Imperfect","score":0.4627000093460083},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4410000145435333},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4350999891757965},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.4178999960422516},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4171999990940094}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7943999767303467},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.61080002784729},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6080999970436096},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5303999781608582},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.46540001034736633},{"id":"https://openalex.org/C2780310539","wikidata":"https://www.wikidata.org/wiki/Q12547192","display_name":"Imperfect","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4499000012874603},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4410000145435333},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4350999891757965},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.4178999960422516},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4171999990940094},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.36570000648498535},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.36160001158714294},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.33009999990463257},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3237000107765198},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3158000111579895},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.2937999963760376},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.27390000224113464},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.2551000118255615}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.02040","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02040","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.02040","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02040","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":"Preprint"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.46385735273361206}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Chain-of-thought":[0],"(CoT)":[1],"reasoning":[2,40,124,200],"has":[3],"significantly":[4],"improved":[5],"the":[6,43,67,77,89,99,118,131,138,158,165,169,174,190,198],"performance":[7,188],"of":[8,45,168,177],"large":[9],"multimodal":[10],"models":[11],"in":[12],"language-guided":[13],"segmentation,":[14],"yet":[15],"its":[16,122],"prohibitive":[17],"computational":[18],"cost,":[19],"stemming":[20],"from":[21,32,206],"generating":[22,98],"verbose":[23],"rationales,":[24],"limits":[25],"real-world":[26],"applicability.":[27],"We":[28],"introduce":[29],"WISE":[30,55],"(Wisdom":[31],"Internal":[33],"Self-Exploration),":[34],"a":[35,57,61,64,72,94,107,126,148,154],"novel":[36],"paradigm":[37],"for":[38,50,53,97],"efficient":[39],"guided":[41],"by":[42,106,202],"principle":[44],"\\textit{thinking":[46],"twice":[47],"--":[48,205],"once":[49,52],"learning,":[51],"speed}.":[54],"trains":[56],"model":[58,119],"to":[59,86,120,208],"generate":[60],"structured":[62],"sequence:":[63],"concise":[65,78,90,171],"rationale,":[66],"final":[68,162],"answer,":[69],"and":[70,115],"then":[71],"detailed":[73,100,123,132],"explanation.":[74,101],"By":[75],"placing":[76],"rationale":[79,91],"first,":[80],"our":[81,143,178],"method":[82],"leverages":[83],"autoregressive":[84],"conditioning":[85],"enforce":[87],"that":[88,110,152,183],"acts":[92],"as":[93],"sufficient":[95],"summary":[96],"This":[102,161],"structure":[103],"is":[104,134,213],"reinforced":[105],"self-distillation":[108],"objective":[109],"jointly":[111],"rewards":[112],"semantic":[113],"fidelity":[114],"conciseness,":[116],"compelling":[117],"internalize":[121],"into":[125,157],"compact":[127],"form.":[128],"At":[129],"inference,":[130],"explanation":[133],"omitted.":[135],"To":[136],"address":[137],"resulting":[139],"conditional":[140],"distribution":[141],"shift,":[142],"inference":[144],"strategy,":[145],"WISE-S,":[146],"employs":[147],"simple":[149],"prompting":[150],"technique":[151],"injects":[153],"brevity-focused":[155],"instruction":[156],"user's":[159],"query.":[160],"adjustment":[163],"facilitates":[164],"robust":[166],"activation":[167],"learned":[170],"policy,":[172],"unlocking":[173],"full":[175],"benefits":[176],"framework.":[179],"Extensive":[180],"experiments":[181],"show":[182],"WISE-S":[184],"achieves":[185],"state-of-the-art":[186],"zero-shot":[187],"on":[189],"ReasonSeg":[191],"benchmark":[192],"with":[193],"58.3":[194],"cIoU,":[195],"while":[196],"reducing":[197],"average":[199],"length":[201],"nearly":[203],"\\textbf{5$\\times$}":[204],"112":[207],"just":[209],"23":[210],"tokens.":[211],"Code":[212],"available":[214],"at":[215],"\\href{https://github.com/mrazhou/WISE}{WISE}.":[216]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-04T00:00:00"}