Artificial Intelligence #llm#reasoning
Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
Researchers propose Minimal Test-Time Intervention (MTI), a training-free method that enhances large language model reasoning by focusing on localized, high-entropy tokens. MTI achieves +9.28% average improvement on six benchmarks for DeepSeek-R1-7B and +11.25% on AIME2024 for Ling-mini-2.0, with minimal computational cost.
Jun 16, 2026 1 source