Bezen

Your daily digest

Research papers, made simple to understand

The AI and machine learning papers people are reading most, each one rewritten in plain words: the problem, the idea, what was found and why it matters.

Safety & Interpretability1 paper, newest first

    • LLM Reasoning
    • Safety & Interpretability

    LLMs often continue explicit reasoning despite instructions to answer directly

    This paper defines and measures "no-thinking" in LLMs as the absence of visible explicit inference before the final answer. Across six LLMs and tasks, explicit no-thinking controls rarely eliminate reasoning traces, especially on open-ended tasks, revealing a trade-off between compliance and accuracy.

    Paper: Thinking Inertia: LLMs Keep Thinking When Told Not To · Dianqiao Lei, Kevin Qinghong Lin, Pan Lu and 2 others

    215 upvotes on Hugging Face13 Oct 20268 min readRead the article