{"id":"https://openalex.org/W7119933689","doi":"https://doi.org/10.48550/arxiv.2601.04765","title":"Differential syntactic and semantic encoding in LLMs","display_name":"Differential syntactic and semantic encoding in LLMs","publication_year":2026,"publication_date":"2026-01-08","ids":{"openalex":"https://openalex.org/W7119933689","doi":"https://doi.org/10.48550/arxiv.2601.04765"},"language":"en","primary_location":{"id":"pmh:oai:repositori.upf.edu:10230/303096","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10230/303096","pdf_url":null,"source":{"id":"https://openalex.org/S4306402615","display_name":"Repositori digital de la UPF (Universitat Pompeu Fabra)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I170486558","host_organization_name":"Universitat Pompeu Fabra","host_organization_lineage":["https://openalex.org/I170486558"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hdl.handle.net/10230/303096","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122445825","display_name":"Santiago Acevedo","orcid":"https://orcid.org/0000-0002-2383-2374","observed_orcids":["https://orcid.org/0000-0002-2383-2374"]},"institutions":[{"id":"https://openalex.org/I138549579","display_name":"Scuola Internazionale Superiore di Studi Avanzati","ror":"https://ror.org/004fze387","country_code":"IT","type":"education","lineage":["https://openalex.org/I138549579"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Acevedo, Santiago","raw_affiliation_strings":["Scuola Internazionale Superiore di Studi Avanzati (SISSA) , Trieste , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Scuola Internazionale Superiore di Studi Avanzati (SISSA) , Trieste , Italy","institution_ids":["https://openalex.org/I138549579"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066542398","display_name":"Alessandro Laio","orcid":"https://orcid.org/0000-0001-9164-7907","observed_orcids":["https://orcid.org/0000-0001-9164-7907"]},"institutions":[{"id":"https://openalex.org/I138549579","display_name":"Scuola Internazionale Superiore di Studi Avanzati","ror":"https://ror.org/004fze387","country_code":"IT","type":"education","lineage":["https://openalex.org/I138549579"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Laio, Alessandro","raw_affiliation_strings":["Scuola Internazionale Superiore di Studi Avanzati (SISSA) , Trieste , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Scuola Internazionale Superiore di Studi Avanzati (SISSA) , Trieste , Italy","institution_ids":["https://openalex.org/I138549579"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5149525234","display_name":"Marco Baroni","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I11932220","display_name":"Instituci\u00f3 Catalana de Recerca i Estudis Avan\u00e7ats","ror":"https://ror.org/0371hy230","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I11932220"]},{"id":"https://openalex.org/I170486558","display_name":"Universitat Pompeu Fabra","ror":"https://ror.org/04n0g0b29","country_code":"ES","type":"education","lineage":["https://openalex.org/I170486558"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Baroni, Marco","raw_affiliation_strings":["Catalan Institute of Research and Advanced Studies (ICREA) and Universitat Pompeu Fabra (UPF) , Barcelona , Spain"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Catalan Institute of Research and Advanced Studies (ICREA) and Universitat Pompeu Fabra (UPF) , Barcelona , Spain","institution_ids":["https://openalex.org/I11932220","https://openalex.org/I170486558"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":5,"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":0.9657999873161316,"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":0.9657999873161316,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.017899999395012856,"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/T10465","display_name":"Neurobiology of Language and Bilingualism","score":0.003800000064074993,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8679999709129333},{"id":"https://openalex.org/keywords/syntactic-encoding","display_name":"syntactic encoding","score":0.8870000243186951},{"id":"https://openalex.org/keywords/semantic-encoding","display_name":"semantic encoding","score":0.9990000128746033},{"id":"https://openalex.org/keywords/deepseek-v3","display_name":"DeepSeek-V3","score":0.7860000133514404},{"id":"https://openalex.org/keywords/hidden-representations","display_name":"hidden representations","score":0.5979999899864197},{"id":"https://openalex.org/keywords/sentence-embeddings","display_name":"sentence embeddings","score":0.5059999823570251}],"concepts":[{"id":"https://openalex.org/C60048249","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntax","level":2,"score":0.7689999938011169},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.671500027179718},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6367999911308289},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.6069999933242798},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.5961999893188477},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.595300018787384},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5889999866485596},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.427700012922287},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.4230000078678131},{"id":"https://openalex.org/C79880414","wikidata":"https://www.wikidata.org/wiki/Q5275338","display_name":"Differential coding","level":3,"score":0.35199999809265137},{"id":"https://openalex.org/C2986991398","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntactic structure","level":3,"score":0.3465000092983246},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C67277372","wikidata":"https://www.wikidata.org/wiki/Q7449085","display_name":"Semantic role labeling","level":3,"score":0.3181999921798706},{"id":"https://openalex.org/C93226319","wikidata":"https://www.wikidata.org/wiki/Q193137","display_name":"Differential (mechanical device)","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C53893814","wikidata":"https://www.wikidata.org/wiki/Q7378909","display_name":"Rule-based machine translation","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C114408938","wikidata":"https://www.wikidata.org/wiki/Q333373","display_name":"Abstract syntax","level":3,"score":0.2572999894618988},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:repositori.upf.edu:10230/303096","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10230/303096","pdf_url":null,"source":{"id":"https://openalex.org/S4306402615","display_name":"Repositori digital de la UPF (Universitat Pompeu Fabra)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I170486558","host_organization_name":"Universitat Pompeu Fabra","host_organization_lineage":["https://openalex.org/I170486558"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"},{"id":"pmh:oai:arXiv.org:2601.04765","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.04765","pdf_url":"https://arxiv.org/pdf/2601.04765","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2601.04765","doi":"https://doi.org/10.48550/arxiv.2601.04765","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.04765","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:repositori.upf.edu:10230/303096","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10230/303096","pdf_url":null,"source":{"id":"https://openalex.org/S4306402615","display_name":"Repositori digital de la UPF (Universitat Pompeu Fabra)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I170486558","host_organization_name":"Universitat Pompeu Fabra","host_organization_lineage":["https://openalex.org/I170486558"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G434337789","display_name":null,"funder_award_id":"HORIZON2020","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G7604509605","display_name":null,"funder_award_id":"SGR-2021","funder_id":"https://openalex.org/F4320334830","funder_display_name":"Ag\u00e8ncia de Gesti\u00f3 d'Ajuts Universitaris i de Recerca"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320334830","display_name":"Ag\u00e8ncia de Gesti\u00f3 d'Ajuts Universitaris i de Recerca","ror":"https://ror.org/01n4pqe45"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7119933689.pdf","grobid_xml":"https://content.openalex.org/works/W7119933689.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W4405903187","https://openalex.org/W4416146115","https://openalex.org/W6910655072","https://openalex.org/W6947991961","https://openalex.org/W7083439586"],"related_works":["https://openalex.org/W7134806122","https://openalex.org/W4388092770","https://openalex.org/W7139148014","https://openalex.org/W7126424400","https://openalex.org/W7164372804","https://openalex.org/W7204213093","https://openalex.org/W2976040964","https://openalex.org/W7167933007","https://openalex.org/W4405253373","https://openalex.org/W4412889010"],"abstract_inverted_index":{"We":[0,24,89],"study":[1],"how":[2],"syntactic":[3,34,48,60],"and":[4,49,61,73,81,99,103],"semantic":[5,50,62],"information":[6,51,123],"is":[7],"encoded":[8],"in":[9,53,124],"inner":[10],"layer":[11],"representations":[12],"of":[13,31,46,97,117,121],"Large":[14],"Language":[15],"Models":[16],"(LLMs),":[17],"focusing":[18],"on":[19],"the":[20,47,54,93,105],"very":[21],"large":[22],"DeepSeek-V3.":[23],"find":[25,91],"that,":[26],"by":[27],"averaging":[28],"hidden-representation":[29],"vectors":[30,40,66],"sentences":[32],"sharing":[33],"structure":[35],"or":[36],"meaning,":[37],"we":[38],"obtain":[39],"that":[41,79,92,104],"capture":[42],"a":[43],"significant":[44],"proportion":[45],"contained":[52],"representations.":[55,126],"In":[56],"particular,":[57],"subtracting":[58],"these":[59,118],"``centroids''":[63],"from":[64],"sentence":[65],"strongly":[67],"affects":[68],"their":[69],"similarity":[70],"with":[71],"syntactically":[72],"semantically":[74],"matched":[75],"sentences,":[76],"respectively,":[77],"suggesting":[78,114],"syntax":[80,98],"semantics":[82,100],"are,":[83],"at":[84],"least":[85],"partially,":[86],"linearly":[87],"encoded.":[88],"also":[90],"cross-layer":[94],"encoding":[95,116],"profiles":[96],"are":[101],"different,":[102],"two":[106,119],"signals":[107],"can":[108],"to":[109],"some":[110],"extent":[111],"be":[112],"decoupled,":[113],"differential":[115],"types":[120],"linguistic":[122],"LLM":[125]},"counts_by_year":[{"year":2026,"cited_by_count":5}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-01-10T00:00:00"}
