{"id":"https://openalex.org/W4416034873","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.740","title":"LLaVE: Large Language and Vision Embedding Models with Hardness-Weighted Contrastive Learning","display_name":"LLaVE: Large Language and Vision Embedding Models with Hardness-Weighted Contrastive Learning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034873","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.740"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.740","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.740","pdf_url":"https://aclanthology.org/2025.findings-emnlp.740.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.740.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112974306","display_name":"Zhibin Lan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhibin Lan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005680009","display_name":"Liqiang Niu","orcid":"https://orcid.org/0000-0002-3366-9486"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liqiang Niu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024849044","display_name":"Fandong Meng","orcid":"https://orcid.org/0000-0002-8158-2377"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fandong Meng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101853761","display_name":"Jie Zhou","orcid":"https://orcid.org/0009-0004-3384-0556"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jie Zhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5066326238","display_name":"Jinsong Su","orcid":"https://orcid.org/0000-0001-5606-7122"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinsong Su","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.8048,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.91877696,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"13721","last_page":"13735"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.47519999742507935,"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":0.47519999742507935,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.36809998750686646,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.024800000712275505,"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/embedding","display_name":"Embedding","score":0.4408000111579895},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.37599998712539673},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3440000116825104},{"id":"https://openalex.org/keywords/contrastive-analysis","display_name":"Contrastive analysis","score":0.30869999527931213},{"id":"https://openalex.org/keywords/language-acquisition","display_name":"Language acquisition","score":0.25870001316070557}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6294000148773193},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6049000024795532},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5094000101089478},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4408000111579895},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.4088999927043915},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.37599998712539673},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3440000116825104},{"id":"https://openalex.org/C2777629044","wikidata":"https://www.wikidata.org/wiki/Q614959","display_name":"Contrastive analysis","level":2,"score":0.30869999527931213},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C2982736386","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Statistical learning","level":2,"score":0.2549000084400177}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.740","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.740","pdf_url":"https://aclanthology.org/2025.findings-emnlp.740.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},{"id":"pmh:doi:10.48550/arxiv.2503.04812","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.740","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.740","pdf_url":"https://aclanthology.org/2025.findings-emnlp.740.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321878","display_name":"Natural Science Foundation of Fujian Province","ror":null}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034873.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Universal":[0],"multimodal":[1,15,18],"embedding":[2,28,75,179],"models":[3,29],"play":[4],"a":[5,37,66,92,142,164],"critical":[6],"role":[7],"in":[8,42,163],"tasks":[9,162],"such":[10],"as":[11],"interleaved":[12],"image-text":[13,154],"retrieval,":[14],"RAG,":[16],"and":[17,47,98,109,129,167],"clustering.":[19],"However,":[20],"our":[21],"empirical":[22],"results":[23,113],"indicate":[24],"that":[25,71,115,120],"existing":[26],"LMM-based":[27],"trained":[30,152],"with":[31,61],"the":[32,74,102,134],"standard":[33],"InfoNCE":[34],"loss":[35],"exhibit":[36],"high":[38],"degree":[39],"of":[40,94,146],"overlap":[41],"similarity":[43],"distribution":[44],"between":[45],"positive":[46],"negative":[48,56,80],"pairs,":[49],"making":[50],"it":[51,156],"challenging":[52],"to":[53,159,177],"distinguish":[54],"hard":[55],"pairs":[57,81],"effectively.":[58],"To":[59],"deal":[60],"this":[62,88],"issue,":[63],"we":[64,90],"propose":[65],"simple":[67],"yet":[68],"effective":[69],"framework":[70],"dynamically":[72],"improves":[73],"model's":[76],"representation":[77],"learning":[78],"for":[79,175],"based":[82],"on":[83,101,153],"their":[84],"discriminative":[85],"difficulty.":[86],"Within":[87],"framework,":[89],"train":[91],"series":[93],"models,":[95,138],"named":[96],"LLaVE,":[97],"evaluate":[99],"them":[100],"MMEB":[103],"benchmark,":[104],"which":[105],"covers":[106],"4":[107],"meta-tasks":[108],"36":[110],"datasets.":[111],"Experimental":[112],"show":[114],"LLaVE":[116,150],"establishes":[117],"stronger":[118],"baselines":[119],"achieve":[121,168],"state-of-the-art":[122],"(SOTA)":[123],"performance":[124,144],"while":[125,139],"demonstrating":[126,171],"strong":[127,169],"scalability":[128],"efficiency.":[130],"Specifically,":[131],"LLaVE-2B":[132],"surpasses":[133],"previous":[135],"SOTA":[136],"7B":[137],"LLaVE-7B":[140],"achieves":[141],"further":[143],"improvement":[145],"6.2":[147],"points.":[148],"Although":[149],"is":[151],"data,":[155],"can":[157],"generalize":[158],"text-video":[160],"retrieval":[161],"zero-shot":[165],"manner":[166],"performance,":[170],"its":[172],"remarkable":[173],"potential":[174],"transfer":[176],"other":[178],"tasks.":[180]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}