{"id":"https://openalex.org/W4416034866","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.537","title":"Learning to Describe Implicit Changes: Noise-robust Pre-training for Image Difference Captioning","display_name":"Learning to Describe Implicit Changes: Noise-robust Pre-training for Image Difference Captioning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034866","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.537"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.537","pdf_url":"https://aclanthology.org/2025.findings-emnlp.537.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.537.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079044890","display_name":"Zixin Guo","orcid":"https://orcid.org/0000-0002-7088-2331"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zixin Guo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Jiayang Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiayang Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077052515","display_name":"Tzu-Jui Julius Wang","orcid":"https://orcid.org/0000-0002-8840-3079"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tzu-Jui Julius Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049343215","display_name":"Abduljalil Radman","orcid":"https://orcid.org/0000-0002-6317-9752"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abduljalil Radman","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071798070","display_name":"Selen Pehlivan","orcid":"https://orcid.org/0000-0002-8658-8890"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Selen Pehlivan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111346891","display_name":"Min Cao","orcid":"https://orcid.org/0000-0002-6628-0370"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Min Cao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036133390","display_name":"Jorma Laaksonen","orcid":"https://orcid.org/0000-0001-7218-3131"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jorma Laaksonen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.9349,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.80793571,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"10125","last_page":"10145"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9851999878883362,"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":0.9851999878883362,"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":0.0026000000070780516,"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/T11439","display_name":"Video Analysis and Summarization","score":0.0019000000320374966,"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/closed-captioning","display_name":"Closed captioning","score":0.8428999781608582},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5321000218391418},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.3296000063419342},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.29120001196861267},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.2612000107765198}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8428999781608582},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6315000057220459},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6082000136375427},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5321000218391418},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.45680001378059387},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3555000126361847},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3296000063419342},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.29120001196861267},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.25949999690055847},{"id":"https://openalex.org/C2987933465","wikidata":"https://www.wikidata.org/wiki/Q141130","display_name":"Image manipulation","level":3,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.537","pdf_url":"https://aclanthology.org/2025.findings-emnlp.537.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"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.537","pdf_url":"https://aclanthology.org/2025.findings-emnlp.537.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":[{"id":"https://openalex.org/G6781587187","display_name":null,"funder_award_id":"62476188","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322725","display_name":"China Scholarship Council","ror":"https://ror.org/04atp4p48"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034866.pdf","grobid_xml":"https://content.openalex.org/works/W4416034866.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Image":[0,102],"Difference":[1],"Captioning":[2,86],"(IDC)":[3],"methods":[4],"have":[5],"advanced":[6],"in":[7,31],"highlighting":[8],"subtle":[9],"differences":[10],"between":[11],"similar":[12],"images,":[13],"but":[14,37],"their":[15],"performance":[16,137],"is":[17],"often":[18,43],"constrained":[19],"by":[20],"limited":[21],"training":[22],"data.Using":[23],"Large":[24],"Multimodal":[25],"Models":[26],"(LMMs)":[27],"to":[28,73,97,105,114],"describe":[29],"changes":[30],"image":[32],"pairs":[33],"mitigates":[34],"data":[35,64,79],"limits":[36],"adds":[38],"noise.These":[39],"change":[40,76,117],"descriptions":[41,77],"are":[42],"coarse":[44],"summaries,":[45],"obscuring":[46],"fine":[47],"details":[48],"and":[49,65,85,94,109,134],"hindering":[50],"noise":[51],"detection.In":[52],"this":[53],"work,":[54],"we":[55],"improve":[56],"IDC":[57,121],"with":[58,70,89,138],"a":[59,82],"noiserobust":[60],"approach":[61],"at":[62],"both":[63],"model":[66,88],"levels.We":[67],"use":[68],"LMMs":[69],"structured":[71],"prompts":[72],"generate":[74],"fine-grained":[75],"during":[78],"curation.We":[80],"propose":[81],"Noise-Aware":[83],"Modeling":[84],"(NAMC)":[87],"three":[90],"modules:":[91],"Noise":[92],"Identification":[93],"Masking":[95],"(NIM)":[96],"reduce":[98],"noisy":[99],"correspondences,":[100],"Masked":[101],"Reconstruction":[103],"(MIR)":[104],"correct":[106],"over-masking":[107],"errors,":[108],"Fine-grained":[110],"Description":[111],"Generation":[112],"(FDG)":[113],"produce":[115],"coherent":[116],"descriptions.Experiments":[118],"on":[119,127],"four":[120],"benchmarks":[122],"show":[123],"that":[124],"NAMC,":[125],"pre-trained":[126],"our":[128],"large-scale":[129],"data,":[130],"outperforms":[131],"streamlined":[132],"architectures":[133],"achieves":[135],"competitive":[136],"LLM-finetuned":[139],"methods,":[140],"offering":[141],"better":[142],"inference":[143],"efficiency.":[144]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}