{"id":"https://openalex.org/W4400141986","doi":"https://doi.org/10.48550/arxiv.2406.19320","title":"Efficient World Models with Context-Aware Tokenization","display_name":"Efficient World Models with Context-Aware Tokenization","publication_year":2024,"publication_date":"2024-06-27","ids":{"openalex":"https://openalex.org/W4400141986","doi":"https://doi.org/10.48550/arxiv.2406.19320"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2406.19320","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.19320","pdf_url":"https://arxiv.org/pdf/2406.19320","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2406.19320","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003334803","display_name":"Vincent Micheli","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I114457229","display_name":"University of Geneva","ror":"https://ror.org/01swzsf04","country_code":"CH","type":"education","lineage":["https://openalex.org/I114457229"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Micheli, Vincent","raw_affiliation_strings":["University of Geneva , Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Geneva , Switzerland","institution_ids":["https://openalex.org/I114457229"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008846212","display_name":"Eloi Alonso","orcid":"https://orcid.org/0000-0003-3565-4015","observed_orcids":["https://orcid.org/0000-0003-3565-4015"]},"institutions":[{"id":"https://openalex.org/I114457229","display_name":"University of Geneva","ror":"https://ror.org/01swzsf04","country_code":"CH","type":"education","lineage":["https://openalex.org/I114457229"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Alonso, Eloi","raw_affiliation_strings":["University of Geneva , Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Geneva , Switzerland","institution_ids":["https://openalex.org/I114457229"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5076094010","display_name":"Fran\u00e7ois Fleuret","orcid":"https://orcid.org/0000-0001-9457-7393","observed_orcids":["https://orcid.org/0000-0001-9457-7393"]},"institutions":[{"id":"https://openalex.org/I114457229","display_name":"University of Geneva","ror":"https://ror.org/01swzsf04","country_code":"CH","type":"education","lineage":["https://openalex.org/I114457229"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Fleuret, Fran\u00e7ois","raw_affiliation_strings":["University of Geneva , Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Geneva , Switzerland","institution_ids":["https://openalex.org/I114457229"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I114457229"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":17,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9994999766349792,"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":0.9994999766349792,"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/T11574","display_name":"Artificial Intelligence in Games","score":0.0003000000142492354,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":9.999999747378752e-05,"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/world-models","display_name":"world models","score":0.8700000047683716},{"id":"https://openalex.org/keywords/model-based-reinforcement-learning","display_name":"model-based reinforcement learning","score":0.9539999961853027},{"id":"https://openalex.org/keywords/autoregressive-transformer","display_name":"autoregressive transformers","score":0.9660000205039978}],"concepts":[{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.7713205814361572},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6000949740409851},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5825130939483643},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.32655367255210876},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2748141288757324},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.18935629725456238},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.053035229444503784}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2406.19320","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.19320","pdf_url":"https://arxiv.org/pdf/2406.19320","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2406.19320","doi":"https://doi.org/10.48550/arxiv.2406.19320","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2406.19320","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2406.19320","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.19320","pdf_url":"https://arxiv.org/pdf/2406.19320","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G6308494385","display_name":"Meaningful Human Control of Security Systems \u2013 Aligning International Humanitarian Law with Human Psychology","funder_award_id":"187494","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4400141986.pdf","grobid_xml":"https://content.openalex.org/works/W4400141986.grobid-xml"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W1959608418","https://openalex.org/W2242818861","https://openalex.org/W2781585732","https://openalex.org/W3206395542","https://openalex.org/W3217227398","https://openalex.org/W4303441796","https://openalex.org/W4303441850","https://openalex.org/W4327810158","https://openalex.org/W4387294122","https://openalex.org/W4409147637"],"related_works":["https://openalex.org/W4294435908","https://openalex.org/W7166830089","https://openalex.org/W4407183545","https://openalex.org/W4403364922","https://openalex.org/W4324298583","https://openalex.org/W7133196033","https://openalex.org/W4403443730","https://openalex.org/W4387583283","https://openalex.org/W4399401105","https://openalex.org/W7161745725"],"abstract_inverted_index":{"Scaling":[0],"up":[1],"deep":[2],"Reinforcement":[3],"Learning":[4],"(RL)":[5],"methods":[6],"presents":[7],"a":[8,21,61,65,71,106],"significant":[9],"challenge.":[10],"Following":[11],"developments":[12],"in":[13,26],"generative":[14],"modelling,":[15],"model-based":[16],"RL":[17],"positions":[18],"itself":[19],"as":[20],"strong":[22],"contender.":[23],"Recent":[24],"advances":[25],"sequence":[27],"modelling":[28],"have":[29],"led":[30],"to":[31,44,51,123],"effective":[32],"transformer-based":[33],"world":[34,66,96],"models,":[35],"albeit":[36],"at":[37,112,135],"the":[38,45,91,95,101,110],"price":[39],"of":[40,48,70,94,109,120],"heavy":[41],"computations":[42],"due":[43],"long":[46],"sequences":[47],"tokens":[49],"required":[50],"accurately":[52],"simulate":[53],"environments.":[54],"In":[55,100],"this":[56],"work,":[57],"we":[58],"propose":[59],"$\u0394$-IRIS,":[60],"new":[62,107],"agent":[63],"with":[64,97],"model":[67],"architecture":[68],"composed":[69],"discrete":[72],"autoencoder":[73],"that":[74,85],"encodes":[75],"stochastic":[76],"deltas":[77,88],"between":[78],"time":[79],"steps":[80],"and":[81,133],"an":[82,118],"autoregressive":[83],"transformer":[84],"predicts":[86],"future":[87],"by":[89],"summarizing":[90],"current":[92],"state":[93,108],"continuous":[98],"tokens.":[99],"Crafter":[102],"benchmark,":[103],"$\u0394$-IRIS":[104],"sets":[105],"art":[111],"multiple":[113],"frame":[114],"budgets,":[115],"while":[116],"being":[117],"order":[119],"magnitude":[121],"faster":[122],"train":[124],"than":[125],"previous":[126],"attention-based":[127],"approaches.":[128],"We":[129],"release":[130],"our":[131],"code":[132],"models":[134],"https://github.com/vmicheli/delta-iris.":[136]},"counts_by_year":[{"year":2026,"cited_by_count":10},{"year":2025,"cited_by_count":7}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2024-06-29T00:00:00"}