{"id":"https://openalex.org/W4416035263","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1714","title":"seqBench: A Tunable Benchmark to Quantify Sequential Reasoning Limits of LLMs","display_name":"seqBench: A Tunable Benchmark to Quantify Sequential Reasoning Limits of LLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035263","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1714"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1714","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1714","pdf_url":"https://aclanthology.org/2025.emnlp-main.1714.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.1714.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031925884","display_name":"Mohammad Ramezanali","orcid":"https://orcid.org/0000-0003-0039-2043"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Mohammad Ramezanali","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015162845","display_name":"M. M. Vazifeh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mo Vazifeh","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5021463856","display_name":"Paolo Santi","orcid":"https://orcid.org/0000-0002-8942-8702"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Paolo Santi","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5031925884"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.18269468,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"33759","last_page":"33780"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11010","display_name":"Logic, Reasoning, and Knowledge","score":0.3116999864578247,"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/T11010","display_name":"Logic, Reasoning, and Knowledge","score":0.3116999864578247,"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/T10456","display_name":"Multi-Agent Systems and Negotiation","score":0.09529999643564224,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.08389999717473984,"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/benchmark","display_name":"Benchmark (surveying)","score":0.5742999911308289},{"id":"https://openalex.org/keywords/sequential-estimation","display_name":"Sequential estimation","score":0.41940000653266907},{"id":"https://openalex.org/keywords/sequential-analysis","display_name":"Sequential analysis","score":0.4187000095844269},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.26600000262260437},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.263700008392334}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6019999980926514},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5742999911308289},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4390000104904175},{"id":"https://openalex.org/C86426650","wikidata":"https://www.wikidata.org/wiki/Q7452504","display_name":"Sequential estimation","level":2,"score":0.41940000653266907},{"id":"https://openalex.org/C80478641","wikidata":"https://www.wikidata.org/wiki/Q195771","display_name":"Sequential analysis","level":2,"score":0.4187000095844269},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.34880000352859497},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3352999985218048},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3278999924659729},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2513999938964844}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.1714","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1714","pdf_url":"https://aclanthology.org/2025.emnlp-main.1714.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1714","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1714","pdf_url":"https://aclanthology.org/2025.emnlp-main.1714.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035263.pdf","grobid_xml":"https://content.openalex.org/works/W4416035263.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"introduce":[1],"seqBench,":[2],"a":[3,70,74,98,106,191],"parametrized":[4],"benchmark":[5],"for":[6,165,201],"probing":[7],"sequential":[8,38],"reasoning":[9,120,151,163],"limits":[10],"in":[11,131,160],"Large":[12],"Language":[13],"Models":[14],"(LLMs)":[15],"through":[16],"precise,":[17],"multi-dimensional":[18],"control":[19,114],"over":[20],"several":[21],"key":[22,71,158],"complexity":[23],"dimensions.seqBench":[24],"allows":[25],"systematic":[26],"variation":[27],"of":[28,37,48,118,194],"(1)":[29],"the":[30,35,43,46,52,58,78,83,91,174],"logical":[31,108],"depth,":[32],"defined":[33,81],"as":[34,82,129],"number":[36,47],"actions":[39],"required":[40],"to":[41,64,168,180,189],"solve":[42],"task;":[44],"(2)":[45],"backtracking":[49],"steps":[50],"along":[51],"optimal":[53],"path,":[54],"quantifying":[55],"how":[56],"often":[57],"agent":[59],"must":[60],"revisit":[61],"prior":[62],"states":[63],"satisfy":[65],"deferred":[66],"preconditions":[67],"(e.g.,":[68],"retrieving":[69],"after":[72],"encountering":[73],"locked":[75],"door);":[76],"and(3)":[77],"noise":[79],"ratio,":[80],"ratio":[84],"between":[85],"supporting":[86],"and":[87,126,138,198],"distracting":[88],"facts":[89],"about":[90],"environment.Our":[92],"evaluations":[93],"on":[94,148],"state-of-the-art":[95],"LLMs":[96],"reveal":[97],"universal":[99,123],"failure":[100],"pattern:":[101],"accuracy":[102],"collapses":[103],"exponentially":[104],"beyond":[105],"modelspecific":[107],"depth.Unlike":[109],"existing":[110],"benchmarks,":[111],"seqBench's":[112,149],"fine-grained":[113],"facilitates":[115],"targeted":[116],"analyses":[117],"these":[119],"failures,":[121],"illuminating":[122],"scaling":[124],"laws":[125],"statistical":[127],"limits,":[128],"detailed":[130],"this":[132],"paper":[133],"alongside":[134],"its":[135],"generation":[136],"methodology":[137],"evaluation":[139],"metrics.We":[140],"find":[141],"that":[142],"even":[143],"top-performing":[144],"models":[145],"systematically":[146],"fail":[147],"structured":[150],"tasks":[152],"despite":[153],"minimal":[154],"search":[155],"complexity,":[156],"underscoring":[157],"limitations":[159],"their":[161,195],"commonsense":[162],"capabilities.Designed":[164],"future":[166],"evolution":[167],"keep":[169],"pace":[170],"with":[171],"advancing":[172],"models,":[173],"seqBench":[175],"datasets":[176],"are":[177],"publicly":[178],"released":[179],"spur":[181],"deeper":[182],"scientific":[183],"inquiry":[184],"into":[185],"LLM":[186],"reasoning,":[187],"aiming":[188],"establish":[190],"clearer":[192],"understanding":[193],"true":[196],"potential":[197],"current":[199],"boundaries":[200],"robust":[202],"real-world":[203],"application.":[204]},"counts_by_year":[],"updated_date":"2026-03-13T14:20:09.374765","created_date":"2025-11-08T00:00:00"}