When2Think: Learning Difficulty-Aware Length Control for Efficient Hybrid Reasoning Models
Large Reasoning Models (LRMs) achieve strong performance on complex tasks but exhibit systematic inefficiency: they often overthink easy problems and underthink hard ones. Existing approaches based on uniform length penalties or rigid routing incur an efficiency tax, trading reduced computation on easy instances for accuracy loss on hard instances. We formulate efficient reasoning as an instance-adaptive computation allocation problem and propose When2Think, a post-training framework for hybrid reasoning that dynamically allocates computation based on problem difficulty. Our method introduces Instance-level Difficulty-Aware Control (IDAC), a reward-shaping mechanism that leverages pre-computed reference statistics (accuracy and token usage) to regulate reasoning depth. Combined with verifier-based rewards and batch-wise standardized advantages, IDAC enables stable critic-free optimization without learned reward models or online reference-model queries. When2Think encourages direct answering on easy instances while preserving extended reasoning on hard instances, thereby learning when to use System 1 (NoThink) versus System 2 (Think). Experiments on mathematical benchmarks demonstrate improved accuracy-efficiency trade-offs: on AIME24, Pass@3 increases by 10.0% while token usage is reduced by 27.9% relative to the base model, and on AIME25, When2Think achieves 40.0% Pass@3, outperforming compression and routing-only baselines.
Code (3)
Similar Papers 제목 키워드 기반
Mitigating Overthinking in Large Reasoning Models via Difficulty-aware Reinforcement Learning
Large Reasoning Models (LRMs) achieve explicit chain-of-thought expansion by imitating deep thinking behaviors of humans, demonstrating excellent performance in complex task scenarios. However, the deep-thinking mode oft…
Reinforcement LearningDAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models
Recent advancements in slow-thinking reasoning models have shown exceptional performance in complex reasoning tasks. However, these models often exhibit overthinking-generating redundant reasoning steps for simple proble…
Between Underthinking and Overthinking: An Empirical Study of Reasoning Length and correctness in LLMs
Large language models (LLMs) are increasingly optimized for long reasoning, under the assumption that more reasoning leads to better performance. However, emerging evidence suggests that longer responses can sometimes de…
Humans Disengage, Reasoning Models Persist: Separating Difficulty Registration from Deliberation Allocation
Large reasoning models (LRMs) take longer on harder problems, just as humans do, but that surface similarity hides an opposite pattern within items. When an LRM gets a problem wrong it spends more tokens than when it get…
ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping
Recent advances in multimodal large reasoning models (MLRMs) have substantially improved their ability to solve complex textual and visual tasks. However, these models tend to overthink on simple problems, producing unne…