paper-with-me

홈 › Papers

Let Me Think! A Long Chain-of-Thought Can Be Worth Exponentially Many Short Ones

2025-05-27 · Parsa Mirtaheri, Ezra Edelman, Samy Jelassi, Eran Malach, Enric Boix-Adsera

Inference-time computation has emerged as a promising scaling axis for improving large language model reasoning. However, despite yielding impressive performance, the optimal allocation of inference-time computation remains poorly understood. A central question is whether to prioritize sequential scaling (e.g., longer chains of thought) or parallel scaling (e.g., majority voting across multiple short chains of thought). In this work, we seek to illuminate the landscape of test-time scaling by demonstrating the existence of reasoning settings where sequential scaling offers an exponential advantage over parallel scaling. These settings are based on graph connectivity problems in challenging distributions of graphs. We validate our theoretical findings with comprehensive experiments across a range of language models, including models trained from scratch for graph connectivity with different chain of thought strategies as well as large reasoning models.

📄 PDF Abstract BibTeX arXiv:2505.21825

Code (1)

seyedparsa/let-me-think 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thought

2025-10-28 · Jiachen Zhao, Yiyou Sun, Weiyan Shi, Dawn Song arxiv

Large language models can generate long chain-of-thought (CoT) reasoning, yet prior work suggests that CoT can be post-hoc rationalization rather than a faithful reflection of the computation through explicitly designed …

Rethinking the Chain-of-Thought: The Roles of In-Context Learning and Pre-trained Priors

2025-09-01 · Hao Yang, Zhiyu Yang, Yunjie Zhang, Shanyi Zhu 외 arxiv

Chain-of-Thought reasoning has emerged as a pivotal methodology for enhancing model inference capabilities. Despite growing interest in Chain-of-Thought reasoning, its underlying mechanisms remain unclear. This paper exp…

Prompt EngineeringLogical Reasoning

The Coupling Tax: How Shared Token Budgets Undermine Visible Chain-of-Thought Under Fixed Output Limits

2026-05-08 · Wenhua Nie, Junlin Liu, Jianan Wu, Zijie Meng 외 arxiv

Chain-of-thought reasoning is often treated as a monotone way to improve language-model accuracy by letting a model think longer. We identify a countervailing effect, the coupling tax: when reasoning traces and final ans…

VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoning

2026-03-15 · Chaoyang Wang, Wenrui Bao, Sicheng Gao, Bingxin Xu 외 arxiv

Vision-Language-Action (VLA) models have shown promising capabilities for embodied intelligence, but most existing approaches rely on text-based chain-of-thought reasoning where visual inputs are treated as static contex…

Reinforcement Learning

Latent Reasoning with Supervised Thinking States

2026-02-09 · Ido Amos, Avi Caciularu, Mor Geva, Amir Globerson 외 arxiv

Reasoning with a chain-of-thought (CoT) enables Large Language Models (LLMs) to solve complex tasks but incurs significant inference costs due to the generation of long rationales. We propose Thinking States, a method th…