Lucene search
+L

27 matches found

Packet Storm News
Packet Storm News
added 2026/03/02 12:0 a.m.16 views

TraceGuard: Process-Guided Firewall against Reasoning Backdoors in Large Language Models

The deployment of Large Reasoning Models LRMs in high-stakes decision-making pipelines has introduced a novel and opaque attack surface: reasoning backdoors. In these attacks, the model's intermediate Chain-of-Thought CoT is manipulated to provide a linguistically plausible but logically fallacio...

6AI score
SaveExploits0
Rows per page
Query Builder