{"id":"https://openalex.org/W4416932012","doi":"https://doi.org/10.48550/arxiv.2511.23476","title":"Thinking by Doing: Building Efficient World Model Reasoning in LLMs via Multi-turn Interaction","display_name":"Thinking by Doing: Building Efficient World Model Reasoning in LLMs via Multi-turn Interaction","publication_year":2025,"publication_date":"2025-11-28","ids":{"openalex":"https://openalex.org/W4416932012","doi":"https://doi.org/10.48550/arxiv.2511.23476"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2511.23476","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.23476","pdf_url":"https://arxiv.org/pdf/2511.23476","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2511.23476","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120710812","display_name":"Bao Shu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shu, Bao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Cai, Yan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cai, Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102818288","display_name":"Jianjian Sun","orcid":"https://orcid.org/0000-0001-8877-0402","observed_orcids":["https://orcid.org/0000-0001-8877-0402"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Jianjian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091891330","display_name":"Chunrui Han","orcid":"https://orcid.org/0000-0002-5777-6315","observed_orcids":["https://orcid.org/0000-0002-5777-6315"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Chunrui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033952997","display_name":"En Yu","orcid":"https://orcid.org/0000-0001-6292-6384","observed_orcids":["https://orcid.org/0000-0001-6292-6384"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, En","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhao, Liang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109632338","display_name":"Jingcheng Hu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Jingcheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Yinmin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yinmin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101443089","display_name":"Haibin Lv","orcid":"https://orcid.org/0000-0002-3064-9818","observed_orcids":["https://orcid.org/0000-0002-3064-9818"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lv, Haoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000360196","display_name":"Yuang Peng","orcid":"https://orcid.org/0000-0002-1448-5489","observed_orcids":["https://orcid.org/0000-0002-1448-5489"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Yuang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085468757","display_name":"Zheng Ge","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ge, Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100362465","display_name":"Xiangyu Zhang","orcid":"https://orcid.org/0000-0003-2138-4608","observed_orcids":["https://orcid.org/0000-0003-2138-4608"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xiangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060116992","display_name":"Daxin Jiang","orcid":"https://orcid.org/0000-0002-6657-5806","observed_orcids":["https://orcid.org/0000-0002-6657-5806"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Daxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5078165161","display_name":"Xiangyu Yue","orcid":"https://orcid.org/0000-0002-6887-2046","observed_orcids":["https://orcid.org/0000-0002-6887-2046"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yue, Xiangyu","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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.8105000257492065,"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":0.8105000257492065,"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":0.16779999434947968,"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":0.006300000008195639,"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/llm-based-agents","display_name":"LLM agents","score":0.8569999933242798},{"id":"https://openalex.org/keywords/multi-turn-interaction","display_name":"multi-turn interaction","score":0.9940000176429749},{"id":"https://openalex.org/keywords/active-learning-machine-learning","display_name":"active learning (machine learning)","score":0.6959999799728394},{"id":"https://openalex.org/keywords/reasoning-efficiency","display_name":"reasoning efficiency","score":0.46799999475479126},{"id":"https://openalex.org/keywords/sokoban","display_name":"Sokoban","score":0.7300000190734863},{"id":"https://openalex.org/keywords/maze","display_name":"mazes","score":0.4740000069141388},{"id":"https://openalex.org/keywords/taxis","display_name":"taxis","score":0.5249999761581421}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5817999839782715},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45500001311302185},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.42559999227523804},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.4185999929904938},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.3734999895095825},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.34150001406669617},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.3353999853134155},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.30480000376701355},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.3041999936103821},{"id":"https://openalex.org/C194995250","wikidata":"https://www.wikidata.org/wiki/Q531136","display_name":"Affordance","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C137549413","wikidata":"https://www.wikidata.org/wiki/Q7053127","display_name":"Normative model of decision-making","level":3,"score":0.2906999886035919},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.28769999742507935},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2856000065803528},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.28529998660087585},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.28380000591278076},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.2786000072956085},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.27410000562667847},{"id":"https://openalex.org/C166088908","wikidata":"https://www.wikidata.org/wiki/Q308495","display_name":"Abductive reasoning","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C61272859","wikidata":"https://www.wikidata.org/wiki/Q7834031","display_name":"Transferability","level":3,"score":0.2646999955177307},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C2780210234","wikidata":"https://www.wikidata.org/wiki/Q422638","display_name":"Action plan","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.2522999942302704}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2511.23476","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.23476","pdf_url":"https://arxiv.org/pdf/2511.23476","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2511.23476","doi":"https://doi.org/10.48550/arxiv.2511.23476","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2511.23476","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:2511.23476","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.23476","pdf_url":"https://arxiv.org/pdf/2511.23476","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"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/W4416038820","https://openalex.org/W7117153217","https://openalex.org/W4414889779","https://openalex.org/W7165749951","https://openalex.org/W7117715996","https://openalex.org/W7216147366","https://openalex.org/W7214155520","https://openalex.org/W4415255786","https://openalex.org/W7127888457","https://openalex.org/W7221541004"],"abstract_inverted_index":{"Developing":[0],"robust":[1],"world":[2,50,90,160],"model":[3,10,51,71,77,91,133,161],"reasoning":[4,39,66,92,162,189],"is":[5],"crucial":[6],"for":[7],"large":[8],"language":[9],"(LLM)":[11],"agents":[12],"to":[13,78,109,122,134,180],"plan":[14],"and":[15,64,85,88,113,138,153,176],"interact":[16],"in":[17,167],"complex":[18,181],"environments.":[19],"While":[20],"multi-turn":[21],"interaction":[22,63,118,128],"offers":[23],"a":[24,37,98,168,186],"superior":[25],"understanding":[26],"of":[27,164,188],"environmental":[28,140,146],"dynamics":[29,141],"via":[30],"authentic":[31],"feedback,":[32],"current":[33],"approaches":[34],"often":[35],"impose":[36],"rigid":[38],"process,":[40],"which":[41,68,130],"constrains":[42],"the":[43,70,76,125,132],"model's":[44],"active":[45,65],"learning,":[46],"ultimately":[47],"hindering":[48],"efficient":[49,62,89],"reasoning.":[52],"To":[53],"address":[54],"these":[55],"issues,":[56],"we":[57],"explore":[58],"world-model":[59],"internalization":[60],"through":[61,82],"(WMAct),":[67],"liberates":[69],"from":[72],"structured":[73],"reasoning,":[74],"allowing":[75],"shape":[79],"thinking":[80],"directly":[81],"its":[83,136],"doing,":[84],"achieves":[86],"effective":[87,159],"with":[93],"two":[94],"key":[95],"mechanisms:":[96],"(1)":[97],"reward":[99,104],"rescaling":[100],"mechanism":[101],"adjusting":[102],"outcome":[103],"based":[105],"on":[106,145,150,185],"action":[107],"efficacy":[108],"incentivize":[110],"redundancy":[111],"reduction":[112],"purposeful":[114],"interaction;":[115],"(2)":[116],"an":[117],"frequency":[119],"annealing":[120],"strategy":[121],"progressively":[123],"reduce":[124],"maximum":[126],"allowed":[127],"turns,":[129],"compels":[131],"condense":[135],"learning":[137],"internalize":[139],"rather":[142],"than":[143],"over-relying":[144],"cues.":[147],"Our":[148],"experiments":[149],"Sokoban,":[151],"Maze,":[152],"Taxi":[154],"show":[155],"that":[156,171],"WMAct":[157],"yields":[158],"capable":[163],"resolving":[165],"tasks":[166],"single":[169],"turn":[170],"previously":[172],"required":[173],"multiple":[174],"interactions":[175],"fosters":[177],"strong":[178],"transferability":[179],"environments,":[182],"improving":[183],"performance":[184],"suite":[187],"benchmarks.":[190]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2025-12-03T00:00:00"}