{"id":"https://openalex.org/W7204911514","doi":"https://doi.org/10.48550/arxiv.2608.29665","title":"Evaluating a 4B open-weights local LLM for agentic DFT workflows: a literature reproducibility audit","display_name":"Evaluating a 4B open-weights local LLM for agentic DFT workflows: a literature reproducibility audit","publication_year":2026,"publication_date":"2026-08-30","ids":{"openalex":"https://openalex.org/W7204911514","doi":"https://doi.org/10.48550/arxiv.2608.29665"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.29665","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.29665","pdf_url":"https://arxiv.org/pdf/2608.29665","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","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/2608.29665","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008808025","display_name":"Shambhu Bhandari Sharma","orcid":"https://orcid.org/0000-0002-6003-7425","observed_orcids":["https://orcid.org/0000-0002-6003-7425"]},"institutions":[{"id":"https://openalex.org/I45129253","display_name":"University College London","ror":"https://ror.org/02jx3x895","country_code":"GB","type":"education","lineage":["https://openalex.org/I124357947","https://openalex.org/I45129253"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Sharma, Shambhu Bhandari","raw_affiliation_strings":["University College London , Department of Earth Sciences , London WC1E 6BT , United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University College London , Department of Earth Sciences , London WC1E 6BT , United Kingdom","institution_ids":["https://openalex.org/I45129253"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5008808025"],"corresponding_institution_ids":["https://openalex.org/I45129253"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.002199999988079071,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13904","display_name":"Artificial Intelligence Applications","score":0.0005000000237487257,"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/agentic-workflows","display_name":"agentic workflows","score":0.6240000128746033},{"id":"https://openalex.org/keywords/local-large-language-models","display_name":"local large language models","score":0.6830000281333923},{"id":"https://openalex.org/keywords/qwen3-4b","display_name":"Qwen3-4B","score":0.9470000267028809},{"id":"https://openalex.org/keywords/density-functional-theory","display_name":"density functional theory","score":0.7699999809265137},{"id":"https://openalex.org/keywords/hardware-constraints","display_name":"hardware constraints","score":0.4560000002384186},{"id":"https://openalex.org/keywords/reproducibility-audit","display_name":"reproducibility audit","score":0.718999981880188},{"id":"https://openalex.org/keywords/lattice-parameters","display_name":"lattice parameters","score":0.7269999980926514}],"concepts":[{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.7659000158309937},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6502000093460083},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.45809999108314514},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.439300000667572},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.43209999799728394},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.41370001435279846},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.3833000063896179},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3698999881744385},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.36419999599456787},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3393999934196472},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3375999927520752},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.32589998841285706},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31619998812675476},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.304500013589859},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C110251889","wikidata":"https://www.wikidata.org/wiki/Q1569697","display_name":"Model checking","level":2,"score":0.29919999837875366},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27390000224113464},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2542000114917755}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.29665","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.29665","pdf_url":"https://arxiv.org/pdf/2608.29665","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.29665","doi":"https://doi.org/10.48550/arxiv.2608.29665","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.29665","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.29665","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.29665","pdf_url":"https://arxiv.org/pdf/2608.29665","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204911514.pdf","grobid_xml":"https://content.openalex.org/works/W7204911514.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W2758504606","https://openalex.org/W4229010617","https://openalex.org/W4414939355","https://openalex.org/W4415137654","https://openalex.org/W4417424660","https://openalex.org/W7124772458","https://openalex.org/W7162494302","https://openalex.org/W7169581516","https://openalex.org/W7196924653"],"related_works":["https://openalex.org/W7223094151","https://openalex.org/W7208704601","https://openalex.org/W7160140912","https://openalex.org/W7162494302","https://openalex.org/W4417424660","https://openalex.org/W4417447074","https://openalex.org/W7125266576","https://openalex.org/W7167470105","https://openalex.org/W7206199170","https://openalex.org/W7222852314"],"abstract_inverted_index":{"Agentic":[0],"workflows":[1,232],"in":[2,187],"materials":[3],"science":[4],"relying":[5],"on":[6,136],"hosted":[7],"commercial":[8],"models":[9,226],"face":[10],"severe":[11],"reproducibility,":[12],"economic,":[13],"and":[14,59,72,84,106,167],"data-privacy":[15],"constraints.":[16,38],"To":[17],"explore":[18],"fully":[19],"local":[20],"agentic":[21,231],"science,":[22],"this":[23],"work":[24],"evaluates":[25],"an":[26,31,171],"open-weights":[27,225],"Qwen3:4B":[28],"model":[29],"executing":[30],"autonomous":[32,230],"scientific":[33],"pipeline":[34],"across":[35],"varying":[36],"hardware":[37,142],"Applied":[39],"to":[40,62,88,162,168,194,199],"pentagonal":[41],"two-dimensional":[42],"materials,":[43],"the":[44,98,127,182,196,201,216],"system":[45],"extracts":[46],"parameters":[47,112,120],"from":[48,160],"unstructured":[49],"text,":[50],"translates":[51],"them":[52],"into":[53],"density":[54],"functional":[55],"theory":[56],"(DFT)":[57],"inputs,":[58],"drives":[60],"simulations":[61],"convergence":[63],"under":[64],"a":[65,132,208],"strict":[66],"neurosymbolic":[67],"architecture":[68],"where":[69,215],"agents":[70],"propose":[71],"deterministic":[73,236],"code":[74,237],"disposes.":[75],"The":[76],"workflow":[77,202],"is":[78],"guarded":[79],"by":[80,235],"verbatim":[81],"text":[82],"grounding":[83],"multi-pass":[85],"inference":[86],"unions":[87],"counteract":[89],"hardware-induced":[90],"structural":[91],"collapse.":[92],"Evaluated":[93],"against":[94],"201":[95],"expert":[96],"judgements,":[97],"extractor":[99],"achieves":[100],"95.7%":[101],"precision":[102,117],"(95%":[103],"CI":[104],"90.3-98.1%)":[105],"67.3%":[107],"recall":[108],"(59.8-74.0%),":[109],"ensuring":[110],"extracted":[111],"are":[113,185],"strictly":[114],"factual.":[115],"However,":[116],"identifying":[118],"absent":[119],"does":[121],"not":[122],"exceed":[123],"47.0%,":[124],"establishing":[125,222],"that":[126,223],"measured":[128],"omission":[129],"rate":[130],"constitutes":[131],"loose":[133],"upper":[134],"bound":[135],"true":[137],"literature":[138],"incompleteness.":[139],"Across":[140],"three":[141],"configurations,":[143],"complete":[144],"GPU":[145],"residency":[146],"governs":[147],"extraction":[148],"quality":[149],"more":[150],"fundamentally":[151],"than":[152],"weight":[153],"or":[154],"cache":[155],"precision,":[156],"raising":[157],"Matthews":[158],"correlation":[159],"0.414":[161],"0.530":[163],"at":[164],"fixed":[165],"quantisation":[166],"0.560":[169],"with":[170,207],"unquantised":[172],"cache.":[173],"A":[174],"corpus-scale":[175],"audit":[176],"indicates":[177],"only":[178],"19":[179],"(33.3%)":[180],"of":[181,213],"57":[183],"studies":[184],"reproducible":[186],"principle,":[188],"reporting":[189],"every":[190],"method":[191],"parameter":[192],"needed":[193],"re-initialise":[195],"calculation.":[197],"Driven":[198],"convergence,":[200],"reproduces":[203],"published":[204],"lattice":[205],"constants":[206],"mean":[209],"absolute":[210],"relative":[211],"error":[212],"2.3%":[214],"relaxed":[217],"structure":[218],"retains":[219],"its":[220],"prototype,":[221],"lightweight":[224],"can":[227],"reliably":[228],"drive":[229],"when":[233],"bounded":[234],"gates.":[238]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-02T00:00:00"}