activity
20232026
most citedRethinking Generalizability and Discriminability of Self-Supervised Learning from Evolutionary Game Theory Perspective

2 citations · 8 across the 39 of their papers we have counts for

collaborators
Showing cs.CLShow all

2 papers · 1 filter

cs.CL2026

Causal Front-Door Adjustment for Robust Jailbreak Attacks on LLMs

Yao Zhou, Zeen Song, Wenwen Qiang +4

Jailbreak attacks on safety-aligned LLMs are increasingly effective, yet attack success alone does not reveal whether compliance is driven by preserved task intent, suppressed safe…

cs.CL2025

COPO: Causal-Oriented Policy Optimization for Hallucinations of MLLMs

Peizheng Guo, Jingyao Wang, Wenwen Qiang +3

Despite Multimodal Large Language Models (MLLMs) having shown impressive capabilities, they may suffer from hallucinations. Empirically, we find that MLLMs attend disproportionatel…