{"id":"https://openalex.org/W7127640580","doi":"https://doi.org/10.48550/arxiv.2602.03211","title":"Lookahead Sample Reward Guidance for Test-Time Scaling of Diffusion Models","display_name":"Lookahead Sample Reward Guidance for Test-Time Scaling of Diffusion Models","publication_year":2026,"publication_date":"2026-02-03","ids":{"openalex":"https://openalex.org/W7127640580","doi":"https://doi.org/10.48550/arxiv.2602.03211"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.03211","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125081765","display_name":"Yeongmin Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Yeongmin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125044109","display_name":"Donghyeok Shin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shin, Donghyeok","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083436740","display_name":"Byeonghu Na","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Na, Byeonghu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125040344","display_name":"Minsang Park","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Park, Minsang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125033101","display_name":"Richard Lee Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Richard Lee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125010003","display_name":"Il-Chul Moon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Moon, Il-Chul","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":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.14206489,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.375,"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":0.375,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.20559999346733093,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.11599999666213989,"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/computation","display_name":"Computation","score":0.6733999848365784},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.644599974155426},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.6087999939918518},{"id":"https://openalex.org/keywords/importance-sampling","display_name":"Importance sampling","score":0.5838000178337097},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.5509999990463257},{"id":"https://openalex.org/keywords/solver","display_name":"Solver","score":0.5077000260353088},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.4507000148296356},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.40149998664855957}],"concepts":[{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6733999848365784},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.644599974155426},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6363000273704529},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6087999939918518},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.5838000178337097},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.578499972820282},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.5509999990463257},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.5077000260353088},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.4507000148296356},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.44290000200271606},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.40149998664855957},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.37310001254081726},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.3702000081539154},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3398999869823456},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.30230000615119934},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.298799991607666},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.2784999907016754},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.2759999930858612},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27129998803138733},{"id":"https://openalex.org/C52421305","wikidata":"https://www.wikidata.org/wiki/Q1151499","display_name":"Particle filter","level":3,"score":0.2563000023365021},{"id":"https://openalex.org/C2781395549","wikidata":"https://www.wikidata.org/wiki/Q4680762","display_name":"Adaptive sampling","level":3,"score":0.2547999918460846},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.03211","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.03211","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.03211","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.03211","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Diffusion":[0],"models":[1],"have":[2],"demonstrated":[3],"strong":[4],"generative":[5],"performance;":[6],"however,":[7],"generated":[8],"samples":[9,85],"often":[10],"fail":[11],"to":[12,121],"fully":[13],"align":[14],"with":[15,30,142],"human":[16],"intent.":[17],"This":[18],"paper":[19],"studies":[20],"an":[21,109],"efficient":[22],"test-time":[23],"scaling":[24],"method":[25,139],"for":[26,37,140],"sampling":[27,52,69,107,123],"from":[28,65,86],"regions":[29],"higher":[31],"human-aligned":[32],"reward":[33,42,93],"values.":[34],"Existing":[35],"methods":[36],"computing":[38],"the":[39,74,130,135,148],"expected":[40],"future":[41],"(EFR)":[43],"face":[44],"important":[45],"limitations:":[46],"backward":[47],"rollout":[48],"incurs":[49],"prohibitively":[50],"high":[51],"costs,":[53],"while":[54],"Tweedie-based":[55],"approaches,":[56],"including":[57],"Sequential":[58],"Monte":[59],"Carlo":[60],"and":[61,67,108],"gradient":[62,137],"guidance,":[63],"suffer":[64],"bias":[66],"inherent":[68],"issues.":[70],"We":[71,119,146],"show":[72],"that":[73,112],"EFR":[75],"at":[76,150],"any":[77],"$\\mathbf{x}_t$":[78],"can":[79],"be":[80],"computed":[81],"using":[82],"only":[83],"marginal":[84],"a":[87,104,143],"pre-trained":[88],"diffusion":[89],"model,":[90],"enabling":[91],"closed-form":[92],"guidance":[94,138],"without":[95],"neural":[96],"backpropagation.":[97],"To":[98],"further":[99],"improve":[100],"efficiency,":[101],"we":[102],"introduce":[103],"few-step":[105],"lookahead":[106,117],"accurate":[110],"solver":[111],"guides":[113],"particles":[114],"toward":[115],"high-reward":[116],"samples.":[118],"refer":[120],"this":[122],"scheme":[124],"as":[125,134],"LiDAR":[126,128],"sampling.":[127],"achieves":[129],"same":[131],"GenEval":[132],"performance":[133],"latest":[136],"SDXL":[141],"9.5x":[144],"speedup.":[145],"release":[147],"code":[149],"https://github.com/aailab-kaist/Diffusion-LiDAR-Sampling.":[151]},"counts_by_year":[],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2026-02-06T00:00:00"}