Characterizing, Evaluating, and Optimizing Complex Reasoning

• URL: http://arxiv.org/abs/2602.08498v1
• Date: Mon, 09 Feb 2026 10:51:14 GMT
• Title: Characterizing, Evaluating, and Optimizing Complex Reasoning
• Authors: Haoran Zhang, Yafu Li, Zhi Wang, Zhilin Wang, Shunkai Zhang, Xiaoye Qu, Yu Cheng,
• Abstract summary: Large Reasoning Models increasingly rely on reasoning traces with complex internal structures.<n>Existing work lacks a unified answer to three fundamental questions.<n>What defines high-quality reasoning, how to reliably evaluate long, implicitly structured reasoning traces, and how to use such evaluation signals for reasoning optimization.
• Score: 44.98294610511283
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Reasoning Models (LRMs) increasingly rely on reasoning traces with complex internal structures. However, existing work lacks a unified answer to three fundamental questions: (1) what defines high-quality reasoning, (2) how to reliably evaluate long, implicitly structured reasoning traces, and (3) how to use such evaluation signals for reasoning optimization. To address these challenges, we provide a unified perspective. (1) We introduce the ME$^2$ principle to characterize reasoning quality along macro- and micro-level concerning efficiency and effectiveness. (2) Built on this principle, we model reasoning traces as directed acyclic graphs (DAGs) and develop a DAG-based pairwise evaluation method, capturing complex reasoning structures. (3) Based on this method, we construct the TRM-Preference dataset and train a Thinking Reward Model (TRM) to evaluate reasoning quality at scale. Experiments show that thinking rewards serve as an effective optimization signal. At test time, selecting better reasoning leads to better outcomes (up to 19.3% gain), and during RL training, thinking rewards enhance reasoning and performance (up to 3.9% gain) across diverse tasks.

Related papers

• Adversarial Yet Cooperative: Multi-Perspective Reasoning in Retrieved-Augmented Language Models [72.4149653187766]
  We propose a Reasoner-Verifier framework named Adrialversa Reasoning RAG (ARR)<n>The Reasoner and Verifier engage in reasoning on retrieved evidence and critiquing each other's logic while being guided by process-aware advantage.<n> Experiments on multiple benchmarks demonstrate the effectiveness of our method.
  arXiv  Detail & Related papers  (2026-01-08T06:57:03Z)
• EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language Models [32.041688648831794]
  We introduce EffiReason-Bench, a unified benchmark for rigorous cross-paradigm evaluation of efficient reasoning methods.<n>To enable step-by-step evaluation, we construct verified CoT annotations for CommonsenseQA and LogiQA.<n>We propose the E3-Score, a principled metric inspired by economic trade-off modeling that provides smooth, stable evaluation without discontinuities.
  arXiv  Detail & Related papers  (2025-11-13T11:14:46Z)
• Incentivizing Consistent, Effective and Scalable Reasoning Capability in Audio LLMs via Reasoning Process Rewards [24.40159537923851]
  We develop a principled method for developing robust and scalable reasoning in Audio Large Language Models.<n>We achieve state-of-the-art results on MMAU Test-mini, substantially outperforming Gemini 2.5 Pro and GPT-4o Audio, and near-human-level performance on MMSU reasoning tasks.
  arXiv  Detail & Related papers  (2025-10-23T06:18:10Z)
• From Sufficiency to Reflection: Reinforcement-Guided Thinking Quality in Retrieval-Augmented Reasoning for LLMs [13.410543801811992]
  This paper analyzes existing RAG reasoning models and identifies three main failure patterns.<n>We propose TIRESRAG-R1, a novel framework using a think-retrieve-reflect process and a multi-dimensional reward system.<n>Experiments on four multi-hop QA datasets show that TIRESRAG-R1 outperforms prior RAG methods and generalizes well to single-hop tasks.
  arXiv  Detail & Related papers  (2025-07-30T14:29:44Z)
• Direct Reasoning Optimization: LLMs Can Reward And Refine Their Own Reasoning for Open-Ended Tasks [6.881699020319577]
  We propose Direct Reasoning Optimization (DRO), a reinforcement learning framework for fine-tuning Large Language Models (LLMs)<n>DRO is guided by a new reward signal: the Reasoning Reflection Reward (R3)<n>DRO consistently outperforms strong baselines while remaining broadly applicable across both open-ended and structured domains.
  arXiv  Detail & Related papers  (2025-06-16T10:43:38Z)
• Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models [68.96619605651155]
  Large reasoning models (LRMs) may drastically increase the output length due to overthinking.<n>We propose a dynamic optimization framework that segments model-generated reasoning paths into distinct thinking patterns.<n>Our method achieves up to a 12% accuracy improvement and reducing token usage from approximately 5,000 to 3,000 tokens.
  arXiv  Detail & Related papers  (2025-05-27T20:59:29Z)
• Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models [49.61246073215651]
  Large Language Models (LLMs) have demonstrated remarkable capabilities in complex tasks.<n>Recent advancements in OpenAI o1 and DeepSeek-R1 have further improved performance in System-2 reasoning domains.<n>However, they also introduce significant computational overhead due to verbose and redundant outputs.
  arXiv  Detail & Related papers  (2025-03-20T17:59:38Z)
• Reward Models Identify Consistency, Not Causality [54.987590763737145]
  State-of-the-art reward models prioritize structural consistency over causal correctness.<n>Removing the problem statement has minimal impact on reward scores.<n> altering numerical values or disrupting the reasoning flow significantly affects RM outputs.
  arXiv  Detail & Related papers  (2025-02-20T14:57:14Z)
• Learning to Plan & Reason for Evaluation with Thinking-LLM-as-a-Judge [78.28188747489769]
  We propose EvalPlanner, a preference optimization algorithm for Thinking-LLM-as-a-Judge.<n>In a self-training loop, EvalPlanner iteratively optimize over synthetically constructed evaluation plans and executions.<n>Our method achieves a new state-of-the-art performance for generative reward models on RewardBench.
  arXiv  Detail & Related papers  (2025-01-30T02:21:59Z)
• RL-STaR: Theoretical Analysis of Reinforcement Learning Frameworks for Self-Taught Reasoner [2.5903660653548366]
  Self-taught reasoner (STaR) uses reinforcement learning to automatically generate reasoning steps.<n> STaR and its variants have demonstrated empirical success, but a theoretical foundation explaining these improvements is lacking.<n>This work provides a theoretical framework for understanding the effectiveness of reinforcement learning on CoT reasoning and STaR.
  arXiv  Detail & Related papers  (2024-10-31T13:17:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.