Research map / Large language models / Language, safety and interpretability
Machine-Translated Knowledge Transfer for Commonsense Causal Reasoning
Jinyoung Yeo, Geungyu Wang, Hyunsouk Cho, Seungtaek Choi, Seung-won Hwang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 PSG (pseudo-parallel alignment and score propagation) to one-best MT with CausalNet | TransQA · COPA Korean | accuracy: .642 -> .690 | Improved |
| add PSG to TransQA | TransQA · COPA Chinese | accuracy: .634 -> .662 | Improved |
| add PSG to TransQA | TransQA · COPA French | accuracy: .604 -> .662 | Improved |
| add synonym expansion (top-3 synonymous causal pairs) to one-best MT | TransQA · COPA Korean | accuracy: .642 -> .546 | Hurt |
| add synonym expansion to TransQA | TransQA · COPA French | accuracy: .604 -> .580 | Hurt |
| increase pseudo-parallel corpus size used for word alignment (ablation) | PSG · COPA Korean and Chinese | accuracy: R-squared 0.91 (Korean), 0.92 (Chinese) | Improved |
| run more causality propagation iterations (about 9) (ablation) | PSG · COPA Korean and Chinese | accuracy | Improved |
| replace 'not A' in machine translation with antonym B | PSG · COPA Korean | accuracy: +0.010 (Korean), +0.014 (Chinese), +0.004 (French) | Improved |
| vary alignment confidence threshold theta (ablation) | PSG · COPA Korean and Chinese | accuracy | No clear effect |
Closest papers by meaning
- Unveiling Causal Reasoning in Large Language Models: Reality or Mirage? (NeurIPS 2024 · similarity 0.89)
- COLD: Causal reasOning in cLosed Daily activities (NeurIPS 2024 · similarity 0.88)
- Language Agents Meet Causality -- Bridging LLMs and Causal World Models (ICLR 2025 · similarity 0.87)
- Nuance Matters: Probing Epistemic Consistency in Causal Reasoning (AAAI 2025 · similarity 0.87)
- Can Large Language Models Infer Causation from Correlation? (ICLR 2024 · similarity 0.87)
- Causality can systematically address the monsters under the bench(marks) (NeurIPS 2026 · similarity 0.86)
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.
