{"id":"https://openalex.org/W7160943424","doi":"https://doi.org/10.48550/arxiv.2605.09346","title":"RuPLaR : Efficient Latent Compression of LLM Reasoning Chains with Rule-Based Priors From Multi-Step to One-Step","display_name":"RuPLaR : Efficient Latent Compression of LLM Reasoning Chains with Rule-Based Priors From Multi-Step to One-Step","publication_year":2026,"publication_date":"2026-05-10","ids":{"openalex":"https://openalex.org/W7160943424","doi":"https://doi.org/10.48550/arxiv.2605.09346"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.09346","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09346","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.09346","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135913246","display_name":"Xiaocheng Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Xiaocheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135934160","display_name":"Kang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Kang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102665698","display_name":"Zaifu Zhan","orcid":"https://orcid.org/0009-0007-5973-2432"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhan, Zaifu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135918263","display_name":"Yuechi Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yuechi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135973236","display_name":"Xiangyu Duan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duan, 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/T10028","display_name":"Topic Modeling","score":0.44589999318122864,"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/T10028","display_name":"Topic Modeling","score":0.44589999318122864,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.10000000149011612,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.08820000290870667,"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/interpretability","display_name":"Interpretability","score":0.9017000198364258},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.6916000247001648},{"id":"https://openalex.org/keywords/constraint","display_name":"Constraint (computer-aided design)","score":0.487199991941452},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.4855000078678131},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.4627000093460083},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.39640000462532043},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.3959999978542328}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.9017000198364258},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.754800021648407},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.6916000247001648},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6144000291824341},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.487199991941452},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.4855000078678131},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4700999855995178},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.39640000462532043},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3959999978542328},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.382999986410141},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.3174000084400177},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.31470000743865967},{"id":"https://openalex.org/C65965080","wikidata":"https://www.wikidata.org/wiki/Q1806885","display_name":"Latent variable model","level":3,"score":0.28529998660087585},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2822999954223633},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2793999910354614}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.09346","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09346","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.09346","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09346","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"Chain-of-Thought":[1,25],"(CoT)":[2],"paradigm,":[3],"while":[4],"enhancing":[5],"the":[6,16,146],"interpretability":[7],"of":[8,21],"Large":[9],"Language":[10],"Models":[11],"(LLMs),":[12],"is":[13],"constrained":[14],"by":[15,97,160],"inefficiencies":[17],"and":[18,56,106,138,178],"expressive":[19],"limits":[20],"natural":[22],"language.":[23],"Latent":[24,71],"(latent":[26],"CoT)":[27],"reasoning,":[28],"which":[29],"operates":[30],"in":[31,46,91,145],"a":[32,37,66,92,115,140],"continuous":[33],"latent":[34,88,164],"space,":[35],"offers":[36],"promising":[38],"alternative":[39],"but":[40,167],"faces":[41],"challenges":[42],"from":[43],"structural":[44],"complexities":[45],"existing":[47,163],"multi-step":[48],"or":[49],"multi-model":[50],"paradigms,":[51],"such":[52],"as":[53],"error":[54],"propagation":[55],"coordination":[57],"overhead.":[58],"In":[59],"this":[60,78,170],"paper,":[61],"we":[62,113],"introduce":[63],"One-Model":[64],"One-Step,":[65],"novel":[67],"compression":[68,154],"framework":[69,155],"for":[70],"Reasoning":[72],"with":[73,128,171],"Rule-Based":[74],"Priors(RuPLaR)":[75],"to":[76,85],"address":[77],"challenge.":[79],"Our":[80],"method":[81],"trains":[82],"an":[83],"LLM":[84],"autonomously":[86],"generate":[87],"reasoning":[89,111],"tokens":[90,127],"single":[93],"training":[94,117],"stage,":[95],"guided":[96],"rule-based":[98,129],"prior":[99],"probability":[100],"distributions,":[101],"thereby":[102],"eliminating":[103],"cascaded":[104],"processes":[105],"inter-model":[107],"dependencies.":[108],"To":[109],"ensure":[110],"quality,":[112],"design":[114],"joint":[116],"objective":[118],"that":[119,152],"enforces":[120],"answer":[121],"consistency":[122],"via":[123,131],"cross-entropy,":[124],"aligns":[125],"soft":[126],"priors":[130],"KL":[132],"divergence":[133],"(the":[134],"Soft":[135],"Thinking":[136],"constraint),":[137],"adds":[139],"problem-thought":[141],"semantic":[142],"alignment":[143],"constraint":[144],"representation":[147],"space.":[148],"Extensive":[149],"experiments":[150],"show":[151],"our":[153],"not":[156],"only":[157],"improves":[158],"accuracy":[159],"11.1%":[161],"over":[162],"CoT":[165],"methods":[166],"also":[168],"achieves":[169],"minimal":[172],"token":[173],"usage,":[174],"underscoring":[175],"its":[176],"effectiveness":[177],"extensibility.":[179],"Code:":[180],"https://github.com/xiaocen-luo/RuPLaR.":[181]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}