{"id":"https://openalex.org/W4417279911","doi":"https://doi.org/10.48550/arxiv.2503.10061","title":"Compute Optimal Scaling of Skills: Knowledge vs Reasoning","display_name":"Compute Optimal Scaling of Skills: Knowledge vs Reasoning","publication_year":2025,"publication_date":"2025-03-13","ids":{"openalex":"https://openalex.org/W4417279911","doi":"https://doi.org/10.48550/arxiv.2503.10061"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2503.10061","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.10061","pdf_url":"https://arxiv.org/pdf/2503.10061","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,"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2503.10061","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112494018","display_name":"Nicholas Roberts","orcid":"https://orcid.org/0009-0003-4762-2202"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Roberts, Nicholas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033738047","display_name":"Niladri S. Chatterji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chatterji, Niladri","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079540764","display_name":"Sharan Narang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Narang, Sharan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Lewis, Mike","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lewis, Mike","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5018069499","display_name":"Dieuwke Hupkes","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hupkes, Dieuwke","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/T10260","display_name":"Software Engineering Research","score":0.21089999377727509,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10260","display_name":"Software Engineering Research","score":0.21089999377727509,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.10019999742507935,"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/T11719","display_name":"Data Quality and Management","score":0.051899999380111694,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.8324999809265137},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.6101999878883362},{"id":"https://openalex.org/keywords/ask-price","display_name":"Ask price","score":0.5896000266075134},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5519000291824341},{"id":"https://openalex.org/keywords/scaling-law","display_name":"Scaling law","score":0.5139999985694885},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.4763000011444092}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.8324999809265137},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6101999878883362},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.5896000266075134},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5519000291824341},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5178999900817871},{"id":"https://openalex.org/C2988430800","wikidata":"https://www.wikidata.org/wiki/Q428971","display_name":"Scaling law","level":3,"score":0.5139999985694885},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.4763000011444092},{"id":"https://openalex.org/C91682802","wikidata":"https://www.wikidata.org/wiki/Q620538","display_name":"Multidimensional scaling","level":2,"score":0.3481000065803528},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34040001034736633},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3336000144481659},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.32820001244544983},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32170000672340393},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.2964000105857849},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.27230000495910645},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2646999955177307},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.260699987411499}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2503.10061","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.10061","pdf_url":"https://arxiv.org/pdf/2503.10061","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,"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.2503.10061","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.10061","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2503.10061","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.10061","pdf_url":"https://arxiv.org/pdf/2503.10061","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,"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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Scaling":[0],"laws":[1,75],"are":[2,76],"a":[3,14,30,132,138],"critical":[4],"component":[5],"of":[6,35,87,96,122],"the":[7,72,88],"LLM":[8],"development":[9],"pipeline,":[10],"most":[11],"famously":[12],"as":[13,21,60],"way":[15],"to":[16,79,127],"forecast":[17],"training":[18],"decisions":[19],"such":[20,59],"'compute-optimally'":[22],"trading-off":[23],"parameter":[24,145],"count":[25,146],"and":[26,56,63,66,99,109,135],"dataset":[27],"size,":[28],"alongside":[29],"more":[31],"recent":[32],"growing":[33],"list":[34],"other":[36],"crucial":[37],"decisions.":[38],"In":[39,51],"this":[40,69],"work,":[41],"we":[42,53,67,91],"ask":[43],"whether":[44,81],"compute-optimal":[45,129,144],"scaling":[46,74,83,115,130],"behaviour":[47],"can":[48,142],"be":[49],"skill-dependent.":[50,77],"particular,":[52],"examine":[54],"knowledge":[55,108],"reasoning-based":[57],"skills":[58],"knowledge-based":[61],"QA":[62],"code":[64,110],"generation,":[65],"answer":[68],"question":[70],"in":[71,114],"affirmative:":[73],"Next,":[78],"understand":[80],"skill-dependent":[82],"is":[84],"an":[85,93,120],"artefact":[86],"pretraining":[89],"datamix,":[90],"conduct":[92],"extensive":[94],"ablation":[95],"different":[97],"datamixes":[98],"find":[100,136],"that,":[101],"also":[102],"when":[103],"correcting":[104],"for":[105],"datamix":[106],"differences,":[107],"exhibit":[111],"fundamental":[112],"differences":[113],"behaviour.":[116],"We":[117],"conclude":[118],"with":[119],"analysis":[121],"how":[123],"our":[124],"findings":[125],"relate":[126],"standard":[128],"using":[131],"validation":[133,140],"set,":[134],"that":[137],"misspecified":[139],"set":[141],"impact":[143],"by":[147],"nearly":[148],"50%,":[149],"depending":[150],"on":[151],"its":[152],"skill":[153],"composition.":[154]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-10-10T00:00:00"}