{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/capturing-row-and-column-semantics-in","title":"Capturing Row and Column Semantics in Transformer Based Question Answering over Tables","arxiv_id":"2104.08303","date":"2021-04-16","proceeding":"NAACL 2021 4","authors":["Michael Glass","Mustafa Canim","Alfio Gliozzo","Saneem Chemmengath","Vishwajeet Kumar","Rishav Chakravarti","Avi Sil","Feifei Pan","Samarth Bharadwaj","Nicolas Rodolfo Fauceglia"],"abstract":"Transformer based architectures are recently used for the task of answering questions over tables. In order to improve the accuracy on this task, specialized pre-training techniques have been developed and applied on millions of open-domain web tables. In this paper, we propose two novel approaches demonstrating that one can achieve superior performance on table QA task without even using any of these specialized pre-training techniques. The first model, called RCI interaction, leverages a transformer based architecture that independently classifies rows and columns to identify relevant cells. While this model yields extremely high accuracy at finding cell values on recent benchmarks, a second model we propose, called RCI representation, provides a significant efficiency advantage for online QA systems over tables by materializing embeddings for existing tables. Experiments on recent benchmarks prove that the proposed methods can effectively locate cell values on tables (up to ~98% Hit@1 accuracy on WikiSQL lookup questions). Also, the interaction model outperforms the state-of-the-art transformer based approaches, pre-trained on very large table corpora (TAPAS and TaBERT), achieving ~3.4% and ~18.86% additional precision improvement on the standard WikiSQL benchmark.","url_abs":"https://arxiv.org/abs/2104.08303v2","url_pdf":"https://arxiv.org/pdf/2104.08303v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"capturing-row-and-column-semantics-in","repo_url":"https://github.com/IBM/row-column-intersection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2104.08303","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2104.08303"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/google-research/tapas","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/IBM/row-column-intersection","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":1,"unverified":9},"by_repo_kind":{"official":{"samples":10,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"7a1a277b5455acbd","entry":"avg","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/aggregation_integration.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/aggregation_integration.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7a1a277b5455acbd"}},{"code_sha256_prefix":"67478c039bf8b9c0","entry":"convert_to_float","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/tapas_eval.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/tapas_eval.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"67478c039bf8b9c0"}},{"code_sha256_prefix":"3e6f5c4103652755","entry":"get_numeric_col_vals","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/aggregation_integration.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/aggregation_integration.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3e6f5c4103652755"}},{"code_sha256_prefix":"4b0deab5f5cb8d50","entry":"log_softmax","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/seq_pair_apply2tables.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/seq_pair_apply2tables.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4b0deab5f5cb8d50"}},{"code_sha256_prefix":"e8e58c66ec0109a2","entry":"sentence_to_inputs","repo":"IBM/row-column-intersection","repo_kind":"official","path":"dataloader/distloader_base.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/dataloader/distloader_base.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e8e58c66ec0109a2"}},{"code_sha256_prefix":"d994de88d7cad170","entry":"standard_json_mapper","repo":"IBM/row-column-intersection","repo_kind":"official","path":"dataloader/distloader_seq_pair.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/dataloader/distloader_seq_pair.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d994de88d7cad170"}},{"code_sha256_prefix":"1be60c6154b0426b","entry":"to_ndarray","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/seq_pair_apply2tables.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/seq_pair_apply2tables.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1be60c6154b0426b"}},{"code_sha256_prefix":"19b6aef837a2c77c","entry":"to_number","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/aggregation_integration.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/aggregation_integration.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"19b6aef837a2c77c"}},{"code_sha256_prefix":"84b82f831409eb64","entry":"tokenize_string","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/tapas_eval.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/tapas_eval.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"84b82f831409eb64"}},{"code_sha256_prefix":"cd5b71c63186c124","entry":"wtq_normalize","repo":"IBM/row-column-intersection","repo_kind":"official","path":"tableqa/tapas_eval.py","file_url":"https://github.com/IBM/row-column-intersection/blob/HEAD/tableqa/tapas_eval.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cd5b71c63186c124"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}