{"id":"https://openalex.org/W7166799345","doi":"https://doi.org/10.18653/v1/2026.acl-long.1658","title":"TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction","display_name":"TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166799345","doi":"https://doi.org/10.18653/v1/2026.acl-long.1658"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1658","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1658","pdf_url":"https://aclanthology.org/2026.acl-long.1658.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.1658.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134845729","display_name":"Chengye Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengye Wang","raw_affiliation_strings":["Yale University Z Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yale University Z Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139797577","display_name":"Lin Fu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin Fu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134874847","display_name":"Zexi Kuang","orcid":null},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zexi Kuang","raw_affiliation_strings":["Yale University Z Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yale University Z Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139773668","display_name":"Yilun Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yilun Zhao","raw_affiliation_strings":["Yale University Z Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yale University Z Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.83026662,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"35821","last_page":"35845"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13523","display_name":"Mathematics, Computing, and Information Processing","score":0.5153999924659729,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T13523","display_name":"Mathematics, Computing, and Information Processing","score":0.5153999924659729,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T12377","display_name":"Digital Humanities and Scholarship","score":0.0812000036239624,"subfield":{"id":"https://openalex.org/subfields/1208","display_name":"Literature and Literary Theory"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12016","display_name":"Web Data Mining and Analysis","score":0.0763000026345253,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.2770000100135803},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.257099986076355},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2500999867916107},{"id":"https://openalex.org/keywords/optical-character-recognition","display_name":"Optical character recognition","score":0.23970000445842743}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6366000175476074},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47040000557899475},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2937000095844269},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.29089999198913574},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.27959999442100525},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.257099986076355},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2500999867916107},{"id":"https://openalex.org/C546480517","wikidata":"https://www.wikidata.org/wiki/Q167555","display_name":"Optical character recognition","level":3,"score":0.23970000445842743},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.23680000007152557}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1658","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1658","pdf_url":"https://aclanthology.org/2026.acl-long.1658.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.1658","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1658","pdf_url":"https://aclanthology.org/2026.acl-long.1658.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166799345.pdf","grobid_xml":"https://content.openalex.org/works/W7166799345.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Existing":[0],"document":[1,103],"OCR":[2],"largely":[3],"targets":[4],"plain":[5],"text":[6],"or":[7],"Markdown,":[8],"discarding":[9],"the":[10],"structural":[11,55,139],"and":[12,31,36,57,71,87,112,120,140],"executable":[13],"properties":[14],"that":[15,50,83,97,126],"make":[16],"LaTeX":[17,30,80],"essential":[18],"for":[19,42],"scientific":[20,26],"publishing.We":[21],"study":[22],"page-level":[23],"reconstruction":[24],"of":[25],"PDFs":[27],"into":[28],"compilable":[29],"introduce":[32],"TEX-OCR-Bench,":[33],"a":[34,38,46,63],"benchmark,":[35],"TEXOCR-Train,":[37,60],"large-scale":[39],"training":[40],"corpus,":[41],"this":[43],"task.TEXOCR-Bench":[44],"features":[45],"multi-dimensional":[47],"evaluation":[48],"suite":[49],"jointly":[51],"assesses":[52],"transcription":[53],"fidelity,":[54],"faithfulness,":[56],"endto-end":[58],"compilability.Leveraging":[59],"we":[61],"train":[62],"2B-parameter":[64],"model,":[65],"TEX-OCR,":[66],"using":[67],"supervised":[68],"fine-tuning":[69],"(SFT)":[70],"reinforcement":[72],"learning":[73],"(RL)":[74],"with":[75,128],"verifiable":[76,129],"rewards":[77,130],"derived":[78],"from":[79],"unit":[81],"tests":[82],"directly":[84],"enforce":[85],"compilability":[86],"referential":[88],"integrity.Experiments":[89],"across":[90],"21":[91],"frontier":[92],"models":[93],"on":[94,138],"TEXOCR-Bench":[95],"show":[96],"existing":[98],"systems":[99],"frequently":[100],"violate":[101],"key":[102],"invariants,":[104],"including":[105],"consistent":[106,132],"section":[107],"structure,":[108],"correct":[109],"float":[110],"placement,":[111],"valid":[113],"label-reference":[114],"links,":[115],"which":[116],"undermines":[117],"compilation":[118,141],"reliability":[119],"downstream":[121],"usability.Our":[122],"analysis":[123],"further":[124],"reveals":[125],"RL":[127],"yields":[131],"improvements":[133],"over":[134],"SFT":[135],"alone,":[136],"particularly":[137],"metrics.":[142]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}