DyCon: Dynamic Reasoning Control via Evolving Difficulty Modeling
Recent advances in Large Reasoning Models (LRMs) demonstrate remarkable performance improvements by iteratively reflecting, exploring, and executing complex tasks, yet suffer from inefficiencies due to redundant reasoning, known as "overthinking". Existing methods to mitigate this issue either rely on static difficulty estimates or require task-specific training, and thus fail to adapt to the dynamic complexity during reasoning. In this work, we empirically show that the problem difficulty evolves dynamically throughout the reasoning process and is linearly encoded in the LRM's step-level embeddings. Building on this insight, we propose DyCon, a training-free framework that leverages latent step-level representations to explicitly model the evolving task difficulty, enabling the dynamic control of reasoning depth to mitigate the overthinking issue. Extensive experiments conducted on four models ranging from 4B to 32B, and across twelve benchmarks in math reasoning, general question answering, and coding tasks demonstrate that DyCon significantly enhances reasoning efficiency by reducing redundant steps without sacrificing accuracy or generalization. Code is available at https://github.com/yu-lin-li/DyCon.
Code (0)
등록된 구현이 없습니다.
Tasks
Question AnsweringSimilar Papers 제목 키워드 기반
DyCON: Dynamic Uncertainty-aware Consistency and Contrastive Learning for Semi-supervised Medical Image Segmentation
Semi-supervised learning in medical image segmentation leverages unlabeled data to reduce annotation burdens through consistency learning. However, current methods struggle with class imbalance and high uncertainty f…
Contrastive LearningImage SegmentationMedical Image SegmentationSemantic Segmentation+1AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting
Modern large reasoning models demonstrate impressive problem-solving capabilities by employing sophisticated reasoning strategies. However, they often struggle to balance efficiency and effectiveness, frequently generati…
GSM8KReinforcement Learning (RL)Group Distributionally Robust Optimization-Driven Reinforcement Learning for LLM Reasoning
Recent progress in Large Language Model (LLM) reasoning is increasingly driven by the refinement of post-training loss functions and alignment strategies. However, standard Reinforcement Learning (RL) paradigms like Grou…
Reinforcement LearningStressEval: Failure-Driven Dynamic Benchmarking for Knowledge-Intensive Reasoning in Large Language Models
Static benchmarks for LLMs are increasingly compromised by contamination and overfitting especially on knowledge intensive reasoning tasks While recent dynamic benchmarks can alleviate staleness they often increase diffi…
MorphoBench: A Benchmark with Difficulty Adaptive to Model Reasoning
With the advancement of powerful large-scale reasoning models, effectively evaluating the reasoning capabilities of these models has become increasingly important. However, existing benchmarks designed to assess the reas…