{"id":"https://openalex.org/W7138202095","doi":"https://doi.org/10.48550/arxiv.2603.15402","title":"A Closer Look into LLMs for Table Understanding","display_name":"A Closer Look into LLMs for Table Understanding","publication_year":2026,"publication_date":"2026-03-16","ids":{"openalex":"https://openalex.org/W7138202095","doi":"https://doi.org/10.48550/arxiv.2603.15402"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.15402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.15402","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.15402","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129687034","display_name":"Jia Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wang, Jia","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129708055","display_name":"Chuanyu Qin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Chuanyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014223736","display_name":"Mingyu Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Mingyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013535642","display_name":"Qingyi Si","orcid":"https://orcid.org/0000-0001-8433-0215"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Si, Qingyi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129694797","display_name":"Peize Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Peize","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129659429","display_name":"Zheng Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Zheng","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5129687034"],"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/T10799","display_name":"Data Visualization and Analytics","score":0.20839999616146088,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10799","display_name":"Data Visualization and Analytics","score":0.20839999616146088,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.10480000078678131,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11516","display_name":"Visual and Cognitive Learning Processes","score":0.06729999929666519,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.8557000160217285},{"id":"https://openalex.org/keywords/interpretability","display_name":"Interpretability","score":0.8251000046730042},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5340999960899353},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.5339999794960022},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.3686999976634979}],"concepts":[{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.8557000160217285},{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8251000046730042},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5340999960899353},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5339999794960022},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4311999976634979},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.3686999976634979},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.31850001215934753},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3149999976158142},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2815000116825104},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.27559998631477356},{"id":"https://openalex.org/C42475967","wikidata":"https://www.wikidata.org/wiki/Q194292","display_name":"Operations research","level":1,"score":0.25870001316070557},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.15402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.15402","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.15402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.15402","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.8187620043754578,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Despite":[0],"the":[1,55,58,62,66,85],"success":[2],"of":[3,68],"Large":[4],"Language":[5],"Models":[6],"(LLMs)":[7],"in":[8,118],"table":[9,86,133],"understanding,":[10],"their":[11,97],"internal":[12],"mechanisms":[13],"remain":[14],"unclear.":[15],"In":[16],"this":[17],"paper,":[18],"we":[19],"conduct":[20],"an":[21],"empirical":[22],"study":[23],"on":[24,51,151],"16":[25],"LLMs,":[26,29,32],"covering":[27],"general":[28],"specialist":[30],"tabular":[31,42,100],"and":[33,44,65,93,123,143,148],"Mixture-of-Experts":[34],"(MoE)":[35],"models,":[36],"to":[37,108],"explore":[38],"how":[39],"LLMs":[40,75],"understand":[41],"data":[43],"perform":[45],"downstream":[46],"tasks.":[47,153],"Our":[48],"analysis":[49],"focus":[50],"4":[52],"dimensions":[53],"including":[54],"attention":[56,79],"dynamics,":[57],"effective":[59],"layer":[60],"depth,":[61],"expert":[63],"activation,":[64],"impacts":[67],"input":[69],"designs.":[70],"Key":[71],"findings":[72,142],"include:":[73],"(1)":[74],"follow":[76],"a":[77],"three-phase":[78],"pattern":[80],"--":[81],"early":[82,122],"layers":[83,89,95,104,125],"scan":[84],"broadly,":[87],"middle":[88,119],"localize":[90],"relevant":[91],"cells,":[92],"late":[94,124],"amplify":[96],"contributions;":[98],"(2)":[99],"tasks":[101],"require":[102],"deeper":[103],"than":[105],"math":[106],"reasoning":[107],"reach":[109],"stable":[110],"predictions;":[111],"(3)":[112],"MoE":[113],"models":[114],"activate":[115],"table-specific":[116],"experts":[117],"layers,":[120],"with":[121],"sharing":[126],"general-purpose":[127],"experts;":[128],"(4)":[129],"Chain-of-Thought":[130],"prompting":[131],"increases":[132],"attention,":[134],"further":[135],"enhanced":[136],"by":[137],"table-tuning.":[138],"We":[139],"hope":[140],"these":[141],"insights":[144],"can":[145],"facilitate":[146],"interpretability":[147],"future":[149],"research":[150],"table-related":[152]},"counts_by_year":[],"updated_date":"2026-03-18T06:31:55.123368","created_date":"2026-03-18T00:00:00"}