{"id":"https://openalex.org/W7166865265","doi":"https://doi.org/10.18653/v1/2026.findings-acl.992","title":"SGG-R 3: From Next-Token Prediction to End-to-End Unbiased Scene Graph Generation","display_name":"SGG-R 3: From Next-Token Prediction to End-to-End Unbiased Scene Graph Generation","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166865265","doi":"https://doi.org/10.18653/v1/2026.findings-acl.992"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.992","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.992","pdf_url":"https://aclanthology.org/2026.findings-acl.992.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: ACL 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-acl.992.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128646214","display_name":"Jiaye Feng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaye Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107662000","display_name":"Qixiang Yin","orcid":"https://orcid.org/0009-0008-7515-1977"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qixiang Yin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101496409","display_name":"Yuankun Liu","orcid":"https://orcid.org/0009-0009-3600-1586"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuankun Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139809707","display_name":"Tong Mo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tong Mo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139750343","display_name":"Weiping Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Weiping Li","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":0.87778695,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"19811","last_page":"19830"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.6614000201225281,"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":0.6614000201225281,"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":0.0681999996304512,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.05270000174641609,"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/graph","display_name":"Graph","score":0.450300008058548},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3287000060081482},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.26980000734329224},{"id":"https://openalex.org/keywords/graph-theory","display_name":"Graph theory","score":0.2540999948978424},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"Image segmentation","score":0.25209999084472656}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5551999807357788},{"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/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.450300008058548},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41190001368522644},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.36480000615119934},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3050000071525574},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.26980000734329224},{"id":"https://openalex.org/C88230418","wikidata":"https://www.wikidata.org/wiki/Q131476","display_name":"Graph theory","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.25209999084472656},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.992","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.992","pdf_url":"https://aclanthology.org/2026.findings-acl.992.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: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.992","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.992","pdf_url":"https://aclanthology.org/2026.findings-acl.992.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: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7998994922","display_name":null,"funder_award_id":"2023YFC3304902","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166865265.pdf","grobid_xml":"https://content.openalex.org/works/W7166865265.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Scene":[0],"Graph":[1],"Generation":[2],"(SGG)":[3],"structures":[4],"visual":[5],"scenes":[6],"as":[7],"graphs":[8,47],"of":[9,31,38,154,182],"objects":[10],"and":[11,35,52,75,112,141,156,180],"their":[12],"relations.While":[13],"Multimodal":[14],"Large":[15],"Language":[16],"Models":[17],"(MLLMs)":[18],"have":[19],"advanced":[20],"end-to-end":[21,94],"SGG,":[22],"current":[23],"methods":[24],"are":[25],"hindered":[26],"by":[27,49,108],"both":[28],"a":[29,63,104,122,134],"lack":[30],"task-specific":[32,69],"structured":[33,64],"reasoning":[34,65,129],"the":[36,99,127,147,178,183],"challenges":[37],"sparse,":[39],"long-tailed":[40],"relation":[41,105,120,143,158],"distributions,":[42],"resulting":[43],"in":[44,88],"incomplete":[45],"scene":[46,96],"characterized":[48],"low":[50],"recall":[51],"biased":[53],"predictions.To":[54],"address":[55],"these":[56],"issues,":[57],"we":[58,102,132],"introduce":[59],"SGG-R":[60,168],"3":[61,169],",":[62],"framework":[66],"that":[67,167],"integrates":[68,139],"Chain-of-Thought":[70],"(CoT)-guided":[71],"Supervised":[72],"Fine-Tuning":[73],"(SFT)":[74],"Reinforcement":[76],"Learning":[77],"(RL)":[78],"with":[79],"Group":[80],"Sequence":[81],"Policy":[82],"Optimization":[83],"(GSPO),":[84],"designed":[85],"to":[86,92,118,174],"engage":[87],"three":[89],"sequential":[90],"stages":[91],"achieve":[93],"unbiased":[95],"graph":[97],"generation.During":[98],"SFT":[100],"phase,":[101],"propose":[103,133],"augmentation":[106],"strategy":[107],"leveraging":[109],"an":[110],"MLLM":[111],"refined":[113],"via":[114,150],"embedding":[115],"similarity":[116],"filtering":[117],"alleviate":[119],"sparsity.Subsequently,":[121],"stagealigned":[123],"reward":[124,137],"scheme":[125],"optimizes":[126],"procedural":[128],"during":[130],"RL.Specifically,":[131],"novel":[135],"dual-granularity":[136],"which":[138],"fine-grained":[140],"coarse-grained":[142],"rewards,":[144],"simultaneously":[145],"mitigating":[146],"long-tail":[148],"issue":[149],"frequency-based":[151],"adaptive":[152],"weighting":[153],"predicates":[155],"improving":[157],"coverage":[159],"through":[160],"semantic":[161],"clustering.Experiments":[162],"on":[163],"two":[164],"benchmarks":[165],"show":[166],"achieves":[170],"superior":[171],"performance":[172],"compared":[173],"existing":[175],"methods,":[176],"demonstrating":[177],"effectiveness":[179],"generalization":[181],"framework.":[184]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}