{"id":"https://openalex.org/W7131874792","doi":"https://doi.org/10.1109/sii64115.2026.11404492","title":"Rewarding Change Beyond State: Directional VLM Rewards for Sample-Efficient Robot Reinforcement Learning <sup>*</sup>","display_name":"Rewarding Change Beyond State: Directional VLM Rewards for Sample-Efficient Robot Reinforcement Learning <sup>*</sup>","publication_year":2026,"publication_date":"2026-01-11","ids":{"openalex":"https://openalex.org/W7131874792","doi":"https://doi.org/10.1109/sii64115.2026.11404492"},"language":null,"primary_location":{"id":"doi:10.1109/sii64115.2026.11404492","is_oa":false,"landing_page_url":"https://doi.org/10.1109/sii64115.2026.11404492","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/SICE International Symposium on System Integration (SII)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5041332315","display_name":"L. Lundgren","orcid":null},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":true,"raw_author_name":"Linus Lundgren","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Computer Science and Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Computer Science and Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030366997","display_name":"Wenhao Lu","orcid":"https://orcid.org/0000-0002-4842-2400"},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Wenhao Lu","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhitao Liang","orcid":null},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Zhitao Liang","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036155189","display_name":"Z. Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Ze Zhang","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Computer Science and Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Computer Science and Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085467629","display_name":"Karinne Ram\u00edrez-Amaro","orcid":"https://orcid.org/0000-0003-4600-1722"},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Karinne Ramirez-Amaro","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]},{"author_position":"last","author":{"id":null,"display_name":"Emmanuel Dean","orcid":null},"institutions":[{"id":"https://openalex.org/I66862912","display_name":"Chalmers University of Technology","ror":"https://ror.org/040wg7k59","country_code":"SE","type":"education","lineage":["https://openalex.org/I66862912"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Emmanuel Dean","raw_affiliation_strings":["Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96"],"affiliations":[{"raw_affiliation_string":"Chalmers University of Technology,Faculty of Electrical Engineering,Gothenburg,Sweden,SE-412 96","institution_ids":["https://openalex.org/I66862912"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5041332315"],"corresponding_institution_ids":["https://openalex.org/I66862912"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.61884368,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"722","last_page":"728"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.8495000004768372,"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"}},"topics":[{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.8495000004768372,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.014399999752640724,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.011099999770522118,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.6669999957084656},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6309000253677368},{"id":"https://openalex.org/keywords/signal","display_name":"SIGNAL (programming language)","score":0.5435000061988831},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.525600016117096},{"id":"https://openalex.org/keywords/term","display_name":"Term (time)","score":0.48840001225471497},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.46939998865127563},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4178999960422516}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6854000091552734},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6669999957084656},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6309000253677368},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6220999956130981},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.5435000061988831},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.525600016117096},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.48840001225471497},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.46939998865127563},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4499000012874603},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4178999960422516},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.38679999113082886},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.3495999872684479},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3328999876976013},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3294999897480011},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.31859999895095825},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.31700000166893005},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.28850001096725464},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.2605000138282776}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/sii64115.2026.11404492","is_oa":false,"landing_page_url":"https://doi.org/10.1109/sii64115.2026.11404492","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/SICE International Symposium on System Integration (SII)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":2,"referenced_works":["https://openalex.org/W4410360555","https://openalex.org/W4414359780"],"related_works":[],"abstract_inverted_index":{"Sparse":[0],"rewards":[1,87],"are":[2],"a":[3,47,51,90,103,107,127,132,156],"persistent":[4],"bottleneck":[5],"for":[6,163,181],"robotic":[7],"manipulation":[8,143],"with":[9,106,118,126,145,193],"Reinforcement":[10],"Learning":[11],"(RL),":[12],"primarily":[13],"because":[14],"RL":[15],"agents":[16],"must":[17],"discover":[18],"long-horizon,":[19],"multi-step":[20],"behaviors":[21],"while":[22],"receiving":[23],"infrequent":[24],"and":[25,79,114,189,205],"weakly":[26],"informative":[27],"feedback.":[28],"Recent":[29],"work":[30],"uses":[31],"pre-trained":[32],"Vision":[33],"Language":[34],"Models":[35],"(VLMs)":[36],"to":[37,81,159,186],"provide":[38],"dense":[39,199],"per-step":[40],"rewards,":[41,201],"yet":[42],"most":[43],"approaches":[44],"score":[45],"only":[46],"single":[48,133],"image":[49],"against":[50],"goal":[52],"text,":[53],"ignoring":[54],"whether":[55],"the":[56,61,111,120,164,182,194],"recent":[57],"change":[58,101],"actually":[59],"moves":[60],"system":[62],"toward":[63],"success.":[64],"We":[65,136],"argue":[66],"that":[67,93],"this":[68],"omission":[69],"impairs":[70],"exploration":[71,204],"(e.g.,":[72],"goal-like":[73],"detours,":[74],"wrong-way":[75],"progress,":[76],"action":[77],"aliasing)":[78],"propose":[80],"make":[82],"time":[83],"explicit":[84],"in":[85,140,171],"VLM":[86,200],"by":[88],"adding":[89],"directional":[91,122,149,184],"signal":[92,123,130],"evaluates":[94],"short-horizon":[95],"change.":[96],"Concretely,":[97],"we":[98],"pair":[99],"visual":[100],"over":[102],"few":[104],"steps":[105],"text":[108],"description":[109],"of":[110,173],"desired":[112],"change,":[113],"finetune":[115],"lightweight":[116],"heads":[117],"RL;":[119],"resulting":[121],"is":[124],"combined":[125],"standard":[128],"positional":[129,166,195],"into":[131],"shaping":[134,150],"reward.":[135],"evaluated":[137],"our":[138],"approach":[139],"six":[141,174],"MetaWorld":[142],"tasks":[144],"fixed":[146,157],"goals.":[147],"This":[148],"improves":[151],"running":[152],"average":[153],"success":[154],"at":[155],"budget":[158],"78.2%,":[160],"versus":[161],"63.8%":[162],"best-tuned":[165],"baseline":[167],"(improvements":[168],"were":[169],"observed":[170],"five":[172],"tasks).":[175],"Ablations":[176],"identify":[177],"key":[178],"design":[179],"choices":[180],"proposed":[183],"term":[185,196],"be":[187],"effective":[188],"show":[190],"its":[191],"synergy":[192],"when":[197],"supplying":[198],"demonstrating":[202],"improved":[203],"sample":[206],"efficiency.":[207]},"counts_by_year":[],"updated_date":"2026-03-25T23:56:10.502304","created_date":"2026-02-28T00:00:00"}