Research map / Probabilistic and scientific modeling / Causal discovery
A Causal Lens for Controllable Text Generation
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 counterfactual confounder-balancing objectives for z and c (ablation) | Causal conditional generation · Yelp | control accuracy: 91.1 -> 96.3 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal conditional generation · Yelp | bias: 89.2 -> 59.8 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal conditional generation · Bios | control accuracy: 97.3 -> 99.2 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal conditional generation · Bios | bias: 70.1 -> 62.4 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal attribute transfer · Biased Yelp | control accuracy: 75.0 -> 77.0 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal attribute transfer · Biased Yelp | bias: 67.8 -> 61.4 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal attribute transfer · Unbiased Yelp | control accuracy: 87.1 -> 91.9 | Helped |
| Add counterfactual confounder-balancing objectives for z and c (ablation) | Causal attribute transfer · Unbiased Yelp | ref-BLEU: 24.3 -> 25.5 | Helped |
Closest papers by meaning
- Rethinking the Development of Large Language Models from the Causal Perspective: A Legal Text Prediction Case Study (AAAI 2024 · similarity 0.85)
- Gumbel Counterfactual Generation From Language Models (ICLR 2025 · similarity 0.85)
- Successor Features for Efficient Multi-Subject Controlled Text Generation (ICML 2024 · similarity 0.85)
- Causality can systematically address the monsters under the bench(marks) (NeurIPS 2026 · similarity 0.85)
- Controlled Text Generation via Language Model Arithmetic (ICLR 2024 · similarity 0.84)
- SCOPE: A Self-supervised Framework for Improving Faithfulness in Conditional Text Generation (ICLR 2025 · 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.