Research map / Large language models / Alignment and preferences
VORTEX: Aligning Task Utility and Human Preferences Through LLM-Guided Reward Shaping
Guojun Xiong, Milind Tambe
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 a human preference via LLM reward shaping | Base Reward policy · Public health simulation | total utility: 432 (median, base reward) then lower under any preference | Hurt |
| apply Favor LI reward shaping | Base Reward policy · Public health simulation | pull proportion of low-income: 0.22 -> 0.75 | Improved |
| apply one reflection step with verbal feedback | VORTEX · Public health illustrative trajectory | utility: 82.4 -> 80.7 (-1.7) | Hurt |
| apply one reflection step with verbal feedback | VORTEX · Public health illustrative trajectory | coverage: 22.0% -> 27.5% (+5.5%) | Improved |
| replace DLM with VORTEX reflective reward shaping | DLM baseline · Public health, Favor LI and Favor LE | total utility stability | Improved |
| replace DLM with VORTEX reflective reward shaping | DLM baseline · Public health, low-income coverage | coverage ratio | Improved |
Closest papers by meaning
- Adaptive Preference Scaling for Reinforcement Learning with Human Feedback (NeurIPS 2024 · similarity 0.87)
- Beyond Reverse KL: Generalizing Direct Preference Optimization with Diverse Divergence Constraints (ICLR 2024 · similarity 0.86)
- Prompted Policy Search: Reinforcement Learning through Linguistic and Numerical Reasoning in LLMs (NeurIPS 2025 · similarity 0.86)
- Sequential Preference Optimization: Multi-Dimensional Preference Alignment with Implicit Reward Modeling (AAAI 2025 · similarity 0.86)
- AlphaDPO: Adaptive Reward Margin for Direct Preference Optimization (ICML 2025 · similarity 0.86)
- Tool-Augmented Reward Modeling (ICLR 2024 · 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.
