| Table-R1: Region-based Reinforcement Learning for Table Understanding | | 51.96 | | | GPT-4o (TCoT) | 2025-05-18 |
| TableBench: A Comprehensive and Complex Benchmark for Table Question Answering | ✓ Link | 51.32 | | | GPT-4-Turbo (TCoT) | 2024-08-17 |
| TableBench: A Comprehensive and Complex Benchmark for Table Question Answering | ✓ Link | 50.39 | | | GPT-4o (TCoT) | 2024-08-17 |
| Table-R1: Region-based Reinforcement Learning for Table Understanding | | 50.06 | | | TableGPT2-72B (TCoT) | 2025-05-18 |
| Efficient Table QA via TableGrid Navigation and Progressive Inference Prompting | | 48.46 | | | TGN (Qwen3-8B) | 2026-05-18 |
| Efficient Table QA via TableGrid Navigation and Progressive Inference Prompting | | 47.17 | | | ReAct (Qwen3-8B) | 2026-05-18 |
| TableBench: A Comprehensive and Complex Benchmark for Table Question Answering | ✓ Link | 43.85 | | | Llama3.1-70B (TCoT) | 2024-08-17 |
| Table-R1: Region-based Reinforcement Learning for Table Understanding | | 43.48 | | | QWQ-32B (TCoT) | 2025-05-18 |
| Table-R1: Region-based Reinforcement Learning for Table Understanding | | 41.85 | | | Table-R1 (Qwen2-7B, TCoT) | 2025-05-18 |
| Efficient Table QA via TableGrid Navigation and Progressive Inference Prompting | | 41.69 | | | CoT (Qwen3-8B) | 2026-05-18 |
| TableBench: A Comprehensive and Complex Benchmark for Table Question Answering | ✓ Link | 35.29 | | | TableLLM-Llama3.1-8B (TCoT) | 2024-08-17 |
| TableBench: A Comprehensive and Complex Benchmark for Table Question Answering | ✓ Link | 30.85 | | | GPT-3.5-Turbo (TCoT) | 2024-08-17 |
| MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering | ✓ Link | | 0.612 | 0.655 | Claude-3.7-Sonnet | 2026-02-10 |
| MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering | ✓ Link | | 0.556 | 0.595 | GPT-4o | 2026-02-10 |
| MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering | ✓ Link | | 0.451 | 0.482 | MATA (avg. across 10 backbones) | 2026-02-10 |
| MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering | ✓ Link | | 0.440 | 0.483 | cogito-32b | 2026-02-10 |
| MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering | ✓ Link | | 0.163 | 0.195 | qwen2.5-3b | 2026-02-10 |
| GLEAN: Grounded Lightweight Evaluation Anchors for Contamination-Aware Tabular Reasoning | | | 0.105 | 0.122 | TAPEX | 2026-01-22 |
| GLEAN: Grounded Lightweight Evaluation Anchors for Contamination-Aware Tabular Reasoning | | | 0.093 | 0.119 | TAPAS-base | 2026-01-22 |
| GLEAN: Grounded Lightweight Evaluation Anchors for Contamination-Aware Tabular Reasoning | | | 0.090 | 0.121 | TAPAS-large | 2026-01-22 |