{"id":"https://openalex.org/W7140106696","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.304","title":"Compositional Reasoning via Joint Image and Language Decomposition","display_name":"Compositional Reasoning via Joint Image and Language Decomposition","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140106696","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.304"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-eacl.304","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.304","pdf_url":"https://aclanthology.org/2026.findings-eacl.304.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: EACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-eacl.304.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5023899979","display_name":"Dwip Dalal","orcid":"https://orcid.org/0009-0007-4063-1076"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dwip Dalal","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130388423","display_name":"Madhav Kanda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Madhav Kanda","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130392532","display_name":"Zhenhailong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhenhailong Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130397083","display_name":"Heng Ji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heng Ji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5015496512","display_name":"Unnat Jain","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Unnat Jain","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.30683025,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5753","last_page":"5775"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.2565000057220459,"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":0.2565000057220459,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.060100000351667404,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.04650000110268593,"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/joint","display_name":"Joint (building)","score":0.5957000255584717},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.44850000739097595},{"id":"https://openalex.org/keywords/decomposition","display_name":"Decomposition","score":0.40610000491142273},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3192000091075897},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.29269999265670776}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6147000193595886},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.5957000255584717},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5860999822616577},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.44850000739097595},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40619999170303345},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.40610000491142273},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4032999873161316},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3192000091075897},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.29269999265670776},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.29019999504089355},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2870999872684479},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27709999680519104},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.25600001215934753},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.25369998812675476}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-eacl.304","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.304","pdf_url":"https://aclanthology.org/2026.findings-eacl.304.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: EACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-eacl.304","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.304","pdf_url":"https://aclanthology.org/2026.findings-eacl.304.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: EACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140106696.pdf","grobid_xml":"https://content.openalex.org/works/W7140106696.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"reasoning":[1,97],"tasks":[2],"such":[3],"as":[4,27],"visual":[5,16,42],"question":[6],"answering":[7],"(VQA)":[8],"require":[9],"models":[10],"to":[11,56,83],"process":[12],"both":[13,37,99],"language":[14,23],"and":[15,39,47,75,94,105,115],"inputs.However,":[17],"existing":[18],"approaches":[19],"typically":[20],"decompose":[21],"only":[22],"queries,":[24],"treating":[25],"images":[26,38],"monolithic":[28],"inputs.We":[29],"introduce":[30],"REDI,":[31],"a":[32],"framework":[33],"that":[34],"jointly":[35],"decomposes":[36],"questions":[40],"into":[41],"sub-domains":[43,59],"(segmentation,":[44],"material,":[45],"depth,":[46],"color)":[48],"with":[49,67],"corresponding":[50],"sub-questions.REDI":[51],"uses":[52],"an":[53],"MLLM":[54],"orchestrator":[55],"select":[57],"the":[58,85],"required":[60],"for":[61],"each":[62],"query,":[63],"generate":[64],"domain-specific":[65],"sub-questions":[66],"grounded":[68],"object":[69,73],"references":[70],"(via":[71],"shared":[72],"labels),":[74],"fuse":[76],"worker":[77],"outputs":[78],"via":[79],"consistency-aware":[80],"aggregation":[81],"(verify-refineoverride)":[82],"produce":[84],"final":[86],"answer.This":[87],"hierarchical":[88],"multi-agent":[89],"design":[90],"mitigates":[91],"error":[92],"propagation":[93],"improves":[95],"compositional":[96],"across":[98],"open-and":[100],"closed-source":[101],"MLLMs.On":[102],"SEEDBench,":[103],"MMBench,":[104],"CLEVR,":[106],"REDI":[107],"achieves":[108],"absolute":[109],"accuracy":[110],"improvements":[111],"of":[112],"8.9%,":[113],"8.2%,":[114],"16.0%":[116],"over":[117],"base":[118],"model.":[119]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-03-24T00:00:00"}