{"id":"https://openalex.org/W7128727152","doi":"https://doi.org/10.48448/wmy9-0456","title":"FantasyTalking2: Timestep-Layer Adaptive Preference Optimization for Audio-Driven Portrait Animation","display_name":"FantasyTalking2: Timestep-Layer Adaptive Preference Optimization for Audio-Driven Portrait Animation","publication_year":2026,"publication_date":"2026-01-07","ids":{"openalex":"https://openalex.org/W7128727152","doi":"https://doi.org/10.48448/wmy9-0456"},"language":"en","primary_location":{"id":"doi:10.48448/wmy9-0456","doi":"https://doi.org/10.48448/wmy9-0456","is_oa":true,"landing_page_url":"https://doi.org/10.48448/wmy9-0456","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"volume":null,"type":"other","is_traditional":false,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/wmy9-0456","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125741655","display_name":"Association for Artificial Intelligence 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Artificial Intelligence 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125935800","display_name":"Fan Jiang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125758299","display_name":"Wang Qiang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiang, Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125744079","display_name":"Mengchao Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Mengchao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125744694","display_name":"Mu Xu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Mu","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9980000257492065,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9980000257492065,"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/T12290","display_name":"Human Motion and Animation","score":0.00039999998989515007,"subfield":{"id":"https://openalex.org/subfields/1704","display_name":"Computer Graphics and Computer-Aided Design"},"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/T10860","display_name":"Speech and Audio Processing","score":0.00039999998989515007,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/preference-optimization","display_name":"preference optimization","score":0.7039999961853027},{"id":"https://openalex.org/keywords/diffusion-models","display_name":"diffusion models","score":0.7419999837875366},{"id":"https://openalex.org/keywords/lip-synchronization","display_name":"lip synchronization","score":0.7390000224113464},{"id":"https://openalex.org/keywords/video-quality","display_name":"video quality","score":0.4830000102519989}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.733299970626831},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.7084000110626221},{"id":"https://openalex.org/C502989409","wikidata":"https://www.wikidata.org/wiki/Q11425","display_name":"Animation","level":2,"score":0.642799973487854},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6279000043869019},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43650001287460327},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.36980000138282776},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.3619000017642975},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3549000024795532},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.33880001306533813},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3384000062942505},{"id":"https://openalex.org/C2777868144","wikidata":"https://www.wikidata.org/wiki/Q7239817","display_name":"Preference elicitation","level":3,"score":0.3203999996185303},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.289000004529953},{"id":"https://openalex.org/C109747225","wikidata":"https://www.wikidata.org/wiki/Q815758","display_name":"Scarcity","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C69369342","wikidata":"https://www.wikidata.org/wiki/Q1401416","display_name":"Computer animation","level":3,"score":0.2831000089645386},{"id":"https://openalex.org/C172367668","wikidata":"https://www.wikidata.org/wiki/Q6504956","display_name":"Data visualization","level":3,"score":0.26980000734329224},{"id":"https://openalex.org/C48007421","wikidata":"https://www.wikidata.org/wiki/Q676252","display_name":"Motion capture","level":3,"score":0.2578999996185303}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/wmy9-0456","doi":"https://doi.org/10.48448/wmy9-0456","is_oa":true,"landing_page_url":"https://doi.org/10.48448/wmy9-0456","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/wmy9-0456","doi":"https://doi.org/10.48448/wmy9-0456","is_oa":true,"landing_page_url":"https://doi.org/10.48448/wmy9-0456","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7110053609","https://openalex.org/W6948098213","https://openalex.org/W7133329952","https://openalex.org/W4390874181","https://openalex.org/W4393248189","https://openalex.org/W4405033244","https://openalex.org/W4416241944","https://openalex.org/W4413147016","https://openalex.org/W4409369524","https://openalex.org/W7150139624"],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2,158,172,183],"audio-driven":[3],"portrait":[4,117],"animation":[5,118],"have":[6],"demonstrated":[7],"impressive":[8],"capabilities.":[9],"However,":[10],"existing":[11,156],"methods":[12,157],"struggle":[13],"to":[14,35,77],"align":[15],"with":[16,47,57,120,160],"fine-grained":[17,142],"human":[18,95,161],"preferences":[19,126],"across":[20,135,144],"multiple":[21],"dimensions,":[22],"such":[23],"as":[24],"motion":[25,175],"naturalness,":[26,176],"lip-sync":[27,173],"accuracy,":[28,174],"and":[29,50,137,177,186],"visual":[30,178],"quality.":[31],"This":[32],"is":[33],"due":[34],"the":[36,51],"difficulty":[37],"of":[38,53],"optimizing":[39],"among":[40],"competing":[41],"preference":[42,59,96,100,162],"objectives,":[43],"which":[44,131],"often":[45],"conflict":[46],"one":[48],"another,":[49],"scarcity":[52],"large-scale,":[54],"high-quality":[55],"datasets":[56],"multidimensional":[58,84,94,122],"annotations.":[60],"To":[61],"address":[62],"these,":[63],"we":[64,89,103],"first":[65],"introduce":[66],"Talking-Critic,":[67],"a":[68,92,111],"multimodal":[69],"reward":[70,75],"model":[71],"that":[72,152],"learns":[73],"human-aligned":[74],"functions":[76],"quantify":[78],"how":[79],"well":[80],"generated":[81],"videos":[82],"satisfy":[83],"expectations.":[85],"Leveraging":[86],"this":[87],"model,":[88],"curate":[90],"Talking-NSQ,":[91],"large-scale":[93],"dataset":[97],"containing":[98],"410K":[99],"pairs.":[101],"Finally,":[102],"propose":[104],"Timestep-Layer":[105],"adaptive":[106],"multi-expert":[107],"Preference":[108],"Optimization":[109],"(TLPO),":[110],"novel":[112],"framework":[113],"for":[114],"aligning":[115,159],"diffusion-based":[116],"models":[119,171],"fine-grained,":[121],"preferences.":[123],"TLPO":[124,165],"decouples":[125],"into":[127],"specialized":[128],"expert":[129],"modules,":[130],"are":[132],"then":[133],"fused":[134],"timesteps":[136],"network":[138],"layers,":[139],"enabling":[140],"comprehensive,":[141],"enhancement":[143],"all":[145],"dimensions":[146],"without":[147],"mutual":[148],"interference.":[149],"Experiments":[150],"demonstrate":[151],"Talking-Critic":[153],"significantly":[154],"outperforms":[155],"ratings.":[163],"Meanwhile,":[164],"achieves":[166],"substantial":[167],"improvements":[168],"over":[169],"baseline":[170],"quality,":[179],"exhibiting":[180],"superior":[181],"performance":[182],"both":[184],"qualitative":[185],"quantitative":[187],"evaluations.":[188]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-02-13T00:00:00"}