Optical Character Recognition2020en
tabfact
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
test
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | ARTEMIS-DA | 93.1 | Dec 2024 | |
| 2 | Dater | 93 | Jan 2023 | |
| 3 | PASTA | 89.3 | Nov 2022 | |
| 4 | Chain-of-Table | 86.61 | Jan 2024 | |
| 5 | Binder | 86 | Oct 2022 | |
| 6 | Tab-PoT | 85.77 | Jun 2024 | |
| 7 | ReasTAP-Large | 84.9 | Oct 2022 | |
| 8 | TAPEX-Large | 84.2 | Jul 2021 | |
| 9 | T5-3b(UnifiedSKG) | 83.68 | Jan 2022 | |
| 10 | Salience-aware TAPAS | 82.1 | Sep 2021 | |
| 11 | TAPAS-Large classifier with Counterfactual + Synthetic pre-training | 81 | Oct 2020 | |
| 12 | TabSQLify (col+row) | 79.5 | Apr 2024 | |
| 13 | NormTab (Targeted) + SQL | 68.9 | Jun 2024 | |
| 14 | Table-BERT-Horizontal-T+F-Template | 65.12 | Sep 2019 | |
| 15 | BERT classifier w/o Table | 50.5 | Sep 2019 |
val
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | PASTA | 89.2 | Nov 2022 | |
| 2 | ReasTAP-Large | 84.6 | Oct 2022 | |
| 3 | TAPEX-Large | 84.6 | Jul 2021 | |
| 4 | T5-3b(UnifiedSKG) | 83.97 | Jan 2022 | |
| 5 | Salience-aware TAPAS | 82.7 | Sep 2021 | |
| 6 | TAPAS-Large classifier with Counterfactual + Synthetic pre-training | 81 | Oct 2020 | |
| 7 | Table-BERT-Horizontal-T+F-Template | 66.1 | Sep 2019 | |
| 8 | BERT classifier w/o Table | 50.9 | Sep 2019 |
Related Papers14
NormTab: Improving Symbolic Reasoning in LLMs Through Tabular Data Normalization
Jun 2024Models: NormTab (Targeted) + SQL
Efficient Prompting for LLM-based Generative Internet of Things
Jun 2024Models: Tab-PoT
TabSQLify: Enhancing Reasoning Capabilities of LLMs Through Table Decomposition
Apr 2024Models: TabSQLify (col+row)
Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding
Jan 2024Models: Chain-of-Table
ReasTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples
Oct 2022Models: ReasTAP-Large
Binding Language Models in Symbolic Languages
Oct 2022Models: Binder
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
Jan 2022Models: T5-3b(UnifiedSKG)
Table-based Fact Verification with Salience-aware Learning
Sep 2021Models: Salience-aware TAPAS
TAPEX: Table Pre-training via Learning a Neural SQL Executor
Jul 2021Models: TAPEX-Large
Understanding tables with intermediate pre-training
Oct 2020Models: TAPAS-Large classifier with Counterfactual + Synthetic pre-training
TabFact: A Large-scale Dataset for Table-based Fact Verification
Sep 2019Models: Table-BERT-Horizontal-T+F-Template, BERT classifier w/o Table