Process or Result? Manipulated Ending Tokens Can Mislead Reasoning LLMs to Ignore the Correct Reasoning Steps

25 March 2025

Papers citing "Process or Result? Manipulated Ending Tokens Can Mislead Reasoning LLMs to Ignore the Correct Reasoning Steps"

2 / 2 papers shown

Title
Token-Efficient Prompt Injection Attack: Provoking Cessation in LLM Reasoning via Adaptive Token Compression Yu Cui Yujun Cai Y. Wang SILM AAML LRM 32 0 0 29 Apr 2025
Safety in Large Reasoning Models: A Survey Cheng Wang Y. Liu B. Li Duzhen Zhang Z. Li Junfeng Fang LRM 49 1 0 24 Apr 2025