Codesota · Computer Vision · Optical Character Recognition · tabfactTasks/Computer Vision/Optical Character Recognition
Optical Character Recognition · benchmark dataset · 2020 · EN

tabfact.

Dataset from Papers With Code

Submit a result ↵
§ 01 · Leaderboard

Best published scores.

30 results indexed across 2 metrics. Shaded row marks current SOTA; ties broken by submission date.

Primary
accuracy · higher is better
All metrics
test, val
test
22 rows
#ModelOrgSubmittedPaper / codetest
01TabTracer—Feb 2026arxiv94.86
02TableMaster—Jan 2025arxiv94.52
03ARTEMIS-DA—Dec 2024ARTEMIS-DA: An Advanced Reasoning and Transformation Eng…93.10
04Dater—Jan 2023Large Language Models are Versatile Decomposers: Decompo… · code93
05STaR-8B—Nov 2025arxiv92.05
06TableMaster—Jan 2025arxiv-gpt4o-mini90.12
07PASTA—Nov 2022PASTA: Table-Operations Aware Fact Verification via Sent… · code89.30
08T-REX (Phi-4)—Aug 2025arxiv89
09PoTable—Dec 2024arxiv88.93
10Chain-of-Table—Jan 2024Chain-of-Table: Evolving Tables in the Reasoning Chain f… · code86.61
11Binder—Oct 2022Binding Language Models in Symbolic Languages · code86
12Tab-PoT—Jun 2024Efficient Prompting for LLM-based Generative Internet of…85.77
13ReasTAP-Large—Oct 2022ReasTAP: Injecting Table Reasoning Skills During Pre-tra… · code84.90
14TAPEX-Large—Jul 2021TAPEX: Table Pre-training via Learning a Neural SQL Exec… · code84.20
15RePanda—Mar 2025arxiv84.09
16T5-3b(UnifiedSKG)—Jan 2022UnifiedSKG: Unifying and Multi-Tasking Structured Knowle… · code83.68
17Salience-aware TAPAS—Sep 2021Table-based Fact Verification with Salience-aware Learni… · code82.10
18TAPAS-Large classifier with Counterfactual + Synthetic pre-training—Oct 2020Understanding tables with intermediate pre-training · code81
19TabSQLify (col+row)—Apr 2024TabSQLify: Enhancing Reasoning Capabilities of LLMs Thro… · code79.50
20NormTab (Targeted) + SQL—Jun 2024NormTab: Improving Symbolic Reasoning in LLMs Through Ta… · code68.90
21Table-BERT-Horizontal-T+F-Template—Sep 2019TabFact: A Large-scale Dataset for Table-based Fact Veri… · code65.12
22BERT classifier w/o Table—Sep 2019TabFact: A Large-scale Dataset for Table-based Fact Veri… · code50.50
val
8 rows
#ModelOrgSubmittedPaper / codeval
01PASTA—Nov 2022PASTA: Table-Operations Aware Fact Verification via Sent… · code89.20
02TAPEX-Large—Jul 2021TAPEX: Table Pre-training via Learning a Neural SQL Exec… · code84.60
03ReasTAP-Large—Oct 2022ReasTAP: Injecting Table Reasoning Skills During Pre-tra… · code84.60
04T5-3b(UnifiedSKG)—Jan 2022UnifiedSKG: Unifying and Multi-Tasking Structured Knowle… · code83.97
05Salience-aware TAPAS—Sep 2021Table-based Fact Verification with Salience-aware Learni… · code82.70
06TAPAS-Large classifier with Counterfactual + Synthetic pre-training—Oct 2020Understanding tables with intermediate pre-training · code81
07Table-BERT-Horizontal-T+F-Template—Sep 2019TabFact: A Large-scale Dataset for Table-based Fact Veri… · code66.10
08BERT classifier w/o Table—Sep 2019TabFact: A Large-scale Dataset for Table-based Fact Veri… · code50.90
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 04 · Literature

14 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies