5 papers
Decoding Hidden Deception in Reasoning LLMs: Activation Explainers for Deception Auditing
Kexin Chen, Yi Liu, Haonan Zhang +3
As LLMs acquire stronger reasoning capabilities, deceptive behavior becomes an increasingly serious safety concern. Existing deception monitors either score visible transcripts or…
LLM-VA: Resolving the Jailbreak-Overrefusal Trade-off via Vector Alignment
Haonan Zhang, Dongxia Wang, Yi Liu +2
Safety-aligned LLMs suffer from two failure modes: jailbreak (answering harmful inputs) and over-refusal (declining benign queries). Existing vector steering methods adjust the mag…
RP-CATE: Recurrent Perceptron-based Channel Attention Transformer Encoder for Industrial Hybrid Modeling
Haoran Yang, Yinan Zhang, Wenjie Zhang +5
Nowadays, industrial hybrid modeling which integrates both mechanistic modeling and machine learning-based modeling techniques has attracted increasing interest from scholars due t…
ORFuzz: Fuzzing the "Other Side" of LLM Safety -- Testing Over-Refusal
Haonan Zhang, Dongxia Wang, Yi Liu +5
Large Language Models (LLMs) increasingly exhibit over-refusal - erroneously rejecting benign queries due to overly conservative safety measures - a critical functional flaw that u…
Sticking to the Mean: Detecting Sticky Tokens in Text Embedding Models
Kexin Chen, Dongxia Wang, Yi Liu +2
Despite the widespread use of Transformer-based text embedding models in NLP tasks, surprising 'sticky tokens' can undermine the reliability of embeddings. These tokens, when repea…