DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference
Recent reasoning Large Language Models (LLMs) demonstrate remarkable problem-solving abilities but often generate long thinking traces whose utility is unclear. Our work aims to improve their efficiency, enabling them to reach high performance without overthinking. First, we analyze the entropy of token probabilities in reasoning traces. Across three models, we observe a consistent U-shaped entropy pattern: high entropy on easy problems despite high accuracy, low entropy on problems with medium difficulty, and high entropy on hard problems reflecting uncertainty. Specifically, we notice 22--25\% entropy reduction from easy to medium difficulty regions, suggesting an {overthinking} phenomenon on easy instances. Building on these insights, we introduce \textbf{DiffAdapt}, a lightweight framework that selects Easy/Normal/Hard inference strategies per question based on their difficulty and reasoning trace entropy. Each inference strategy consists of a fixed prompt, temperature and maximum token length. In contrast to existing efficiency optimization methods, our approach does not fine-tune base LLM but a small probe that classifies LLM's final hidden state, allowing inexpensive adaptation. We comprehensively evaluate our method on five models and eight benchmarks. Our method achieves comparable or improved accuracy while reducing token usage by up to 22.4\%, establishing a practical path toward compute-efficient reasoning.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
CODA: Difficulty-Aware Compute Allocation for Adaptive Reasoning
The emergence of large reasoning models demonstrates that scaling inference-time compute significantly enhances performance on complex tasks. However, it often falls into another trap: overthinking simple problems, where…
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…
Concise Reasoning, Big Gains: Pruning Long Reasoning Trace with Difficulty-Aware Prompting
Existing chain-of-thought (CoT) distillation methods can effectively transfer reasoning abilities to base models but suffer from two major limitations: excessive verbosity of reasoning traces and inadequate adaptability …
The LLM Already Knows: Estimating LLM-Perceived Question Difficulty via Hidden Representations
Estimating the difficulty of input questions as perceived by large language models (LLMs) is essential for accurate performance evaluation and adaptive inference. Existing methods typically rely on repeated response samp…
DAST: 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…