{"ID":2892414,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.15974","arxiv_id":"2507.15974","title":"Does More Inference-Time Compute Really Help Robustness?","abstract":"Recently, Zaremba et al. demonstrated that increasing inference-time computation improves robustness in large proprietary reasoning LLMs. In this paper, we first show that smaller-scale, open-source models (e.g., DeepSeek R1, Qwen3, Phi-reasoning) can also benefit from inference-time scaling using a simple budget forcing strategy. More importantly, we reveal and critically examine an implicit assumption in prior work: intermediate reasoning steps are hidden from adversaries. By relaxing this assumption, we identify an important security risk, intuitively motivated and empirically verified as an inverse scaling law: if intermediate reasoning steps become explicitly accessible, increased inference-time computation consistently reduces model robustness. Finally, we discuss practical scenarios where models with hidden reasoning chains are still vulnerable to attacks, such as models with tool-integrated reasoning and advanced reasoning extraction attacks. Our findings collectively demonstrate that the robustness benefits of inference-time scaling depend heavily on the adversarial setting and deployment context. We urge practitioners to carefully weigh these subtle trade-offs before applying inference-time scaling in security-sensitive, real-world applications.","short_abstract":"Recently, Zaremba et al. demonstrated that increasing inference-time computation improves robustness in large proprietary reasoning LLMs. In this paper, we first show that smaller-scale, open-source models (e.g., DeepSeek R1, Qwen3, Phi-reasoning) can also benefit from inference-time scaling using a simple budget forci...","url_abs":"https://arxiv.org/abs/2507.15974","url_pdf":"https://arxiv.org/pdf/2507.15974v1","authors":"[\"Tong Wu\",\"Chong Xiang\",\"Jiachen T. Wang\",\"Weichen Yu\",\"Chawin Sitawarin\",\"Vikash Sehwag\",\"Prateek Mittal\"]","published":"2025-07-21T18:08:38Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\"]","has_code":false}