{"id":"https://openalex.org/W7119215832","doi":"https://doi.org/10.48550/arxiv.2601.00647","title":"Physio-DPO: Aligning Large Language Models with the Protein Energy Landscape to Eliminate Structural Hallucinations","display_name":"Physio-DPO: Aligning Large Language Models with the Protein Energy Landscape to Eliminate Structural Hallucinations","publication_year":2026,"publication_date":"2026-01-02","ids":{"openalex":"https://openalex.org/W7119215832","doi":"https://doi.org/10.48550/arxiv.2601.00647"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2601.00647","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2601.00647","pdf_url":"https://arxiv.org/pdf/2601.00647","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","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2601.00647","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076228611","display_name":"Qiwei Meng","orcid":"https://orcid.org/0000-0001-9272-7896","observed_orcids":["https://orcid.org/0000-0001-9272-7896"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Meng, QiWei","raw_affiliation_strings":["Jiaotong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jiaotong University","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5076228611"],"corresponding_institution_ids":[],"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/T10044","display_name":"Protein Structure and Dynamics","score":0.9916999936103821,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T10044","display_name":"Protein Structure and Dynamics","score":0.9916999936103821,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12254","display_name":"Machine Learning in Bioinformatics","score":0.0020000000949949026,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10521","display_name":"RNA and protein synthesis mechanisms","score":0.0017000000225380063,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/protein-energy-landscape","display_name":"protein energy landscape","score":0.6549999713897705},{"id":"https://openalex.org/keywords/direct-preference-optimization","display_name":"direct preference optimization","score":0.7570000290870667},{"id":"https://openalex.org/keywords/protein-design","display_name":"protein design","score":0.6100000143051147},{"id":"https://openalex.org/keywords/thermodynamic-stability","display_name":"thermodynamic stability","score":0.843999981880188},{"id":"https://openalex.org/keywords/hard-negative-sampling","display_name":"hard negative sampling","score":0.550000011920929},{"id":"https://openalex.org/keywords/foldability","display_name":"foldability","score":0.75},{"id":"https://openalex.org/keywords/hydrophobic-core-packing","display_name":"hydrophobic core packing","score":0.8939999938011169},{"id":"https://openalex.org/keywords/hydrogen-bond-network","display_name":"hydrogen bond network","score":0.8299999833106995}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5716000199317932},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.4934000074863434},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.492000013589859},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4722000062465668},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.4025000035762787},{"id":"https://openalex.org/C18051474","wikidata":"https://www.wikidata.org/wiki/Q899656","display_name":"Protein structure prediction","level":3,"score":0.39640000462532043},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C119621388","wikidata":"https://www.wikidata.org/wiki/Q5377166","display_name":"Energy landscape","level":2,"score":0.3743000030517578},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.36090001463890076},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.36059999465942383},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.3555999994277954},{"id":"https://openalex.org/C47701112","wikidata":"https://www.wikidata.org/wiki/Q735188","display_name":"Protein structure","level":2,"score":0.33390000462532043},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33000001311302185},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.3292999863624573},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3190000057220459},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.30559998750686646},{"id":"https://openalex.org/C204328495","wikidata":"https://www.wikidata.org/wiki/Q847556","display_name":"Protein folding","level":2,"score":0.2904999852180481},{"id":"https://openalex.org/C14961307","wikidata":"https://www.wikidata.org/wiki/Q5377176","display_name":"Energy minimization","level":2,"score":0.28940001130104065},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.2838999927043915},{"id":"https://openalex.org/C32598905","wikidata":"https://www.wikidata.org/wiki/Q6497130","display_name":"Lattice protein","level":3,"score":0.2721000015735626},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2628999948501587}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2601.00647","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2601.00647","pdf_url":"https://arxiv.org/pdf/2601.00647","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.2601.00647","doi":"https://doi.org/10.48550/arxiv.2601.00647","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.00647","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:2601.00647","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2601.00647","pdf_url":"https://arxiv.org/pdf/2601.00647","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7119215832.pdf","grobid_xml":"https://content.openalex.org/works/W7119215832.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W2736601468","https://openalex.org/W4391555989","https://openalex.org/W4404782689","https://openalex.org/W4415947883","https://openalex.org/W6891904700"],"related_works":["https://openalex.org/W7203457270","https://openalex.org/W4398163621","https://openalex.org/W4416453581","https://openalex.org/W7203439560","https://openalex.org/W4388099178","https://openalex.org/W4319071809","https://openalex.org/W7155388031","https://openalex.org/W7203889922","https://openalex.org/W4391636496","https://openalex.org/W7128383462"],"abstract_inverted_index":{"Large":[0],"Protein":[1],"Language":[2],"Models":[3],"have":[4],"shown":[5],"strong":[6,105],"potential":[7],"for":[8],"generative":[9],"protein":[10,70],"design,":[11],"yet":[12],"they":[13,44],"frequently":[14],"produce":[15],"structural":[16,133],"hallucinations,":[17],"generating":[18],"sequences":[19],"with":[20],"high":[21],"linguistic":[22],"likelihood":[23],"that":[24,68,82,101,129],"fold":[25],"into":[26],"thermodynamically":[27],"unstable":[28],"conformations.":[29],"Existing":[30],"alignment":[31,66],"approaches":[32],"such":[33,139],"as":[34,43,47,140],"Direct":[35],"Preference":[36],"Optimization":[37],"are":[38],"limited":[39],"in":[40,73],"this":[41],"setting,":[42],"model":[45],"preferences":[46],"binary":[48],"labels":[49],"and":[50,94,110,120,144],"ignore":[51],"the":[52,56,88],"continuous":[53],"structure":[54],"of":[55],"physical":[57],"energy":[58,89],"landscape.":[59],"We":[60],"propose":[61],"Physio-DPO,":[62],"a":[63,78],"physics":[64,95],"informed":[65],"framework":[67],"grounds":[69],"language":[71],"models":[72],"thermodynamic":[74],"stability.":[75],"Physio-DPO":[76,102,130],"introduces":[77],"magnitude":[79],"aware":[80],"objective":[81],"scales":[83],"optimization":[84],"updates":[85],"according":[86],"to":[87,117,123],"gap":[90],"between":[91],"native":[92],"structures":[93],"perturbed":[96],"hard":[97],"negatives.":[98],"Experiments":[99],"show":[100],"consistently":[103],"outperforms":[104],"baselines":[106],"including":[107],"SFT,":[108],"PPO,":[109],"standard":[111],"DPO,":[112],"reducing":[113],"self":[114],"consistency":[115],"RMSD":[116],"1.28":[118],"\u00c5":[119],"increasing":[121],"foldability":[122],"92.8%.":[124],"Qualitative":[125],"analysis":[126],"further":[127],"demonstrates":[128],"effectively":[131],"mitigates":[132],"hallucinations":[134],"by":[135],"recovering":[136],"biophysical":[137],"interactions":[138],"hydrophobic":[141],"core":[142],"packing":[143],"hydrogen":[145],"bond":[146],"networks.":[147]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-01-08T00:00:00"}