# Rethinking Reasoning Quality in Large Language Models through Enhanced Chain-of-Thought via RL

arXiv: https://arxiv.org/abs/2509.06024 · HTML: https://mlsanity.com/paper/2509.06024/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 23 of its 65 references that are in the map.
- First posted on arXiv 7 Sep 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 23 of 23)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://mlsanity.com/paper/2501.12599/) (2025)
- [Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs](https://mlsanity.com/paper/2503.01307/) (2025)
- [ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search](https://mlsanity.com/paper/2406.03816/) (2024)
- [Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought](https://mlsanity.com/paper/2210.01240/) (2022)
- [Transformers as Soft Reasoners over Language](https://mlsanity.com/paper/2002.05867/) (2020)
- [To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning](https://mlsanity.com/paper/2409.12183/) (2024)
- [REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization](https://mlsanity.com/paper/2501.03262/) (2025)
- [FOLIO: Natural Language Reasoning with First-Order Logic](https://mlsanity.com/paper/2209.00840/) (2022)
- [ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning](https://mlsanity.com/paper/2502.01100/) (2025)
- [LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical Reasoning](https://mlsanity.com/paper/2007.08124/) (2020)
- [CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction](https://mlsanity.com/paper/2502.07316/) (2025)
- [Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search](https://mlsanity.com/paper/2502.02508/) (2025)
- [Empowering LLMs with Logical Reasoning: A Comprehensive Survey](https://mlsanity.com/paper/2502.15652/) (2025)
- [Reasoning-Enhanced Healthcare Predictions with Knowledge Graph Community Retrieval](https://mlsanity.com/paper/2410.04585/) (2024)
- [AutoLogi: Automated Generation of Logic Puzzles for Evaluating Reasoning Abilities of Large Language Models](https://mlsanity.com/paper/2502.16906/) (2025)
- [JustLogic: A Comprehensive Benchmark for Evaluating Deductive Reasoning in Large Language Models](https://mlsanity.com/paper/2501.14851/) (2025)
- [LogicGame: Benchmarking Rule-Based Reasoning Abilities of Large Language Models](https://mlsanity.com/paper/2408.15778/) (2024)
- [Step-by-Step Reasoning to Solve Grid Puzzles: Where do LLMs Falter?](https://mlsanity.com/paper/2407.14790/) (2024)

## Built on by (most cited first, 5 of 5)

- [ESTAR: Early-Stopping Token-Aware Reasoning For Efficient Inference](https://mlsanity.com/paper/2602.10004/) (2026)
- [How to Train a Real-World Silicon Concierge? Internalizing Complex Business Workflow to Only OneModel](https://mlsanity.com/paper/2608.20350/) (2026)
- [The Answer Lies Within: Self-Derived Rewards Enable Explainable Relation Extraction](https://mlsanity.com/paper/2510.06198/) (2025)
- [DiPO: Disentangled Perplexity Policy Optimization for Fine-grained Exploration-Exploitation Trade-Off](https://mlsanity.com/paper/2604.13902/) (2026)
- [Ask, Condition or Abstain: Reinforcement Learning for Missing-Premise Reasoning](https://mlsanity.com/paper/2608.16554/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) has recently become the dominant paradigm for strengthening the reasoning abilities of large language models (LLMs). Yet the rule-based reward functions commonly used on mathematical or programming benchmarks assess only answer format and correctness, providing no signal as to whether the induced Chain-of-Thought (CoT) actually improves the answer. Furthermore, such task-specific training offers limited control over logical depth and therefore may fail to reveal a model's genuine reasoning capacity. We propose Dynamic Reasoning Efficiency Reward (DRER) -- a plug-and-play RL reward framework that reshapes both reward and advantage signals. (i) A Reasoning Quality Reward assigns fine-grained credit to those reasoning chains that demonstrably raise the likelihood of the correct answer, directly incentivising the trajectories with beneficial CoT tokens. (ii) A Dynamic Length Advantage decays the advantage of responses whose length deviates from a validation-derived threshold, stabilising training. To facilitate rigorous assessment, we also release Logictree, a dynamically constructed deductive reasoning dataset that functions both as RL training data and as a comprehensive benchmark. Experiments confirm the effectiveness of DRER: our 7B model attains GPT-o3-mini level performance on Logictree with 400 trianing steps, while the average confidence of CoT-augmented answers rises by 30%. The model further exhibits generalisation across diverse logical-reasoning datasets, and the mathematical benchmark AIME24. These results illuminate how RL shapes CoT behaviour and chart a practical path toward enhancing formal-reasoning skills in large language models. All code and data are available in repository https://github.com/Henryhe09/DRER.