Research map / Large language models / Language, safety and interpretability
Table-to-Text: Describing Table Region With Natural Language
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add copying from table cells (ablation) | Table2Seq · WikiTableText | BLEU-4: 5.41 -> 37.90 | Helped |
| Add copying from table cells (ablation) | Table2Seq · WikiBio | BLEU-4: 36.88 -> 40.26 | Helped |
| Add copying from table cells (ablation) | Table2Seq · SimpleQuestions | BLEU-4: 14.70 -> 38.85 | Helped |
| Add caption input (ablation) | Table2Seq · WikiTableText | BLEU-4: 27.06 -> 37.90 | Helped |
| Add global table information (ablation) | Table2Seq · WikiTableText | BLEU-4: 36.68 -> 37.90 | Helped |
| Add local attribute information (ablation) | Table2Seq · WikiTableText | BLEU-4: 36.50 -> 37.90 | Helped |
| Add global table information (ablation) | Table2Seq · WikiBio | BLEU-4: 40.11 -> 40.26 | Hurt |
| Add local attribute information (ablation) | Table2Seq · WikiBio | BLEU-4: 40.03 -> 40.26 | Helped |
| Add global table information (ablation) | Table2Seq · SimpleQuestions | BLEU-4: 39.12 -> 38.85 | Hurt |
| Add local attribute information (ablation) | Table2Seq · SimpleQuestions | BLEU-4: 38.32 -> 38.85 | Helped |
| Extend copying to table attributes | Table2Seq++ · WikiTableText | BLEU-4: 37.90 -> 38.23 | Helped |
| Replace learned copying with random cell copying | Table2Seq · WikiTableText | BLEU-4: 12.01 vs 37.90 | Hurt |
Closest papers by meaning
- NeedleInATable: Exploring Long-Context Capability of Large Language Models towards Long-Structured Tables (NeurIPS 2025 · similarity 0.85)
- UniTabE: A Universal Pretraining Protocol for Tabular Foundation Model in Data Science (ICLR 2024 · similarity 0.85)
- SynTab-LLaVA: Enhancing Multimodal Table Understanding with Decoupled Synthesis (CVPR 2025 · similarity 0.84)
- TableBench: A Comprehensive and Complex Benchmark for Table Question Answering (AAAI 2025 · similarity 0.84)
- Table as a Modality for Large Language Models (NeurIPS 2025 · similarity 0.84)
- TableRAG: Million-Token Table Understanding with Language Models (NeurIPS 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.