# Rewarding Better Thinking for LLM Preference Alignment

arXiv: https://arxiv.org/abs/2607.19824 · HTML: https://mlsanity.com/paper/2607.19824/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 24 of its 39 references that are in the map.
- First posted on arXiv 22 Jul 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 24 of 24)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [Instruction Tuning with GPT-4](https://mlsanity.com/paper/2304.03277/) (2023)
- [DeepSeek LLM: Scaling Open-Source Language Models with Longtermism](https://mlsanity.com/paper/2401.02954/) (2024)
- [Skywork-Reward: Bag of Tricks for Reward Modeling in LLMs](https://mlsanity.com/paper/2410.18451/) (2024)
- [Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning](https://mlsanity.com/paper/2502.14768/) (2025)
- [Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains](https://mlsanity.com/paper/2503.23829/) (2025)
- [OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment](https://mlsanity.com/paper/2510.07743/) (2025)
- [CheckEval: A reliable LLM-as-a-Judge framework for evaluating text generation using checklists](https://mlsanity.com/paper/2403.18771/) (2024)
- [Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Tasks](https://mlsanity.com/paper/2602.05125/) (2026)
- [ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs](https://mlsanity.com/paper/2506.18896/) (2025)
- [ReCode: Reinforcing Code Generation with Reasoning-Process Rewards](https://mlsanity.com/paper/2508.05170/) (2025)
- [Uncertainty-Aware Step-wise Verification with Generative Reward Models](https://mlsanity.com/paper/2502.11250/) (2025)
- [RLSR: Reinforcement Learning from Self Reward](https://mlsanity.com/paper/2505.08827/) (2025)
- [AutoRule: Reasoning Chain-of-thought Extracted Rule-based Rewards Improve Preference Learning](https://mlsanity.com/paper/2506.15651/) (2025)
- [Good Learners Think Their Thinking: Generative PRM Makes Large Reasoning Model More Efficient Math Learner](https://mlsanity.com/paper/2507.23317/) (2025)
- [ReasonGRM: Enhancing Generative Reward Models through Large Reasoning Models](https://mlsanity.com/paper/2506.16712/) (2025)
- [AutoChecklist: Composable Pipelines for Checklist Generation and Scoring with LLM-as-a-Judge](https://mlsanity.com/paper/2603.07019/) (2026)
- [P2S: Probabilistic Process Supervision for General-Domain Reasoning Question Answering](https://mlsanity.com/paper/2601.20649/) (2026)

## Abstract (arXiv)

LLM preference alignment aims to optimize models toward human preferences across diverse user instructions. Reinforcement learning has become a major post-training approach for this goal, but existing proxy rewards are often outcome-level, mainly evaluating the final response while providing limited guidance for the reasoning trajectory. This can make credit assignment coarse when multiple responses receive similar final scores, leaving trajectory-level preferences under-specified. To address this limitation, we propose Thinking Checklist Reward (TCR), a process-oriented reward for RL-based preference alignment. TCR converts preference pairs into sample-specific thinking checklists and uses them to evaluate whether the generated reasoning trace addresses the preference-implied considerations. To reduce overlap with outcome-level supervision, TCR further introduces an exponential moving average (EMA) residual formulation to isolate a complementary thinking surplus beyond what is predictable from the outcome reward. Experiments on five models from three model families show that TCR consistently improves alignment performance across diverse benchmarks, with ablations further validating the importance of EMA-based residual formulation and sample-specific checklist supervision.