Does Thinking More always Help? Understanding Test-Time Scaling in Reasoning Models
Recent trends in test-time scaling for reasoning models (e.g., OpenAI o1, DeepSeek R1) have led to a popular belief that extending thinking traces using prompts like "Wait" or "Let me rethink" can improve performance. This raises a natural question: Does thinking more at test-time truly lead to better reasoning? To answer this question, we perform a detailed empirical study across models and benchmarks, which reveals a consistent pattern of initial performance improvements from additional thinking followed by a decline, due to "overthinking". To understand this non-monotonic trend, we consider a simple probabilistic model, which reveals that additional thinking increases output variance-creating an illusion of improved reasoning while ultimately undermining precision. Thus, observed gains from "more thinking" are not true indicators of improved reasoning, but artifacts stemming from the connection between model uncertainty and evaluation metric. This suggests that test-time scaling through extended thinking is not an effective way to utilize the inference thinking budget. Recognizing these limitations, we introduce an alternative test-time scaling approach, parallel thinking, inspired by Best-of-N sampling. Our method generates multiple independent reasoning paths within the same inference budget and selects the most consistent response via majority vote, achieving up to 20% higher accuracy compared to extended thinking. This provides a simple yet effective mechanism for test-time scaling of reasoning models.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Rethinking Pre-training and Self-training
Pre-training is a dominant paradigm in computer vision. For example, supervised ImageNet pre-training is commonly used to initialize the backbones of object detection and segmentation models. He et al., however, show a s…
Data AugmentationObjectobject-detectionObject Detection+2Towards Understanding and Modeling Empathy for Use in Motivational Design Thinking
Design Thinking workshops are used by companies to help generate new ideas for technologies and products by engaging subjects in exercises to understand their users' wants and become more empathetic towards their needs. …
PositionWhat if...?: Thinking Counterfactual Keywords Helps to Mitigate Hallucination in Large Multi-modal Models
This paper presents a way of enhancing the reliability of Large Multi-modal Models (LMMs) in addressing hallucination, where the models generate cross-modal inconsistent responses. Without additional training, we propose…
counterfactualHallucinationScene UnderstandingThinking Longer, Not Always Smarter: Evaluating LLM Capabilities in Hierarchical Legal Reasoning
Case-based reasoning is a cornerstone of U.S. legal practice, requiring professionals to argue about a current case by drawing analogies to and distinguishing from past precedents. While Large Language Models (LLMs) have…
Legal ReasoningTIME: Temporally Intelligent Meta-reasoning Engine for Context-Triggered Explicit Reasoning
Reasoning-oriented language models typically expose explicit reasoning as a long, front-loaded chain of "thinking" tokens before the main output, either always enabled or externally toggled at inference time. Although th…