Large Language Model (LLM) agents are increasingly acting through Web interfaces and external tools. They produce trajectories. Those trajectories may pass through several intermediaries. They can store, modify, or dispute those trajectories. In Web-based agent deployments, trajectories may pass through APIs, marketpla...
CACHELINT is designed, which models a prompt-caching mis-integration as an invocation site together with the preconditions required for correct cache integration, and locates caching invocation sites with CodeQL and discharges each precondition either against a dated provider capability oracle or with an LLM guard chec...
Ying Zhang, Ruo-Qi Guo, Yi Liu et al.· 0 citations
AdvPCS is proposed, a universal cross-prompt adversarial attack for Promptable Concept Segmentation (PCS), including a min-max prompt optimization strategy, a global-local perception deception attack, and a temporal transition deviation attack.
Zi-Qi Zhou, Yi-Fan Hu, Yu-Fei Song et al.· 0 citations
Watermarking LLM-generated text is an important task for tracing its provenance. Existing LLM watermarks preserve provenance under editing, but this same robustness allows an adversary to alter critical content while retaining attribution, a vulnerability known as piggyback spoofing. We introduce an innovative watermar...
Xiao-Yang Feng, Yanjun Zhang, He Zhang et al.· 0 citations
This work proposes IdentityMask, a robust protection framework that shifts the paradigm from arbitrary confusion to precise, targeted feature corruption, and consistently outperforms prior state-of-the-art approaches in both protection efficacy and robustness.
Wei-Wei Tan, Rui Wang, Lihua Jing et al.· Proceedings of the Thirty-Fi...· 0 citations
JevAdvBench is introduced, to the authors' knowledge the first adversarial benchmark for RLCD models, with 812 typed questions over 66 scenarios, and a black-box attack suite of 9,744 single-edit variants that each edit one part of a request, with billed input tokens confirming that the edit reached the model.
Jian-Yi Hu, Hang Zhang, Yi Liu et al.· 12 citations· ⚡3
ODPure is proposed, a novel input-stage black-box defense for object detection, which is based on input purification that ensures stable perception flows and provides robust defense against diverse backdoor attacks and trigger types while preserving baseline accuracy.
Li Zeng, Ming-Cheng Duan, Long-Fei Fan et al.· 0 citations
RLCDAlignBench is presented, which benchmarks Jev on ten alignment failures: sycophancy, jailbreaks, deception, prompt injection, hallucination, privacy violation, social bias, reward hacking, concealing uncertainty, and power seeking.
Ruo-Qi Guo, Yi Liu, Ge-Lei Deng et al.· 6 citations
The results show that reliable malicious-Skill detection requires both broader cross-source benchmark coverage and evaluation that jointly measures attack detection and benign over-flagging.
Gradient-seeded Reinforcement Learning And Stealthy Pruning (GRASP), a three-stage framework that tackles challenges of adversarial attacks on machine learning-based malware detectors, and out-performs baselines, achieving higher attack success with fewer queries and smaller file-size inflation.
Yu-Tong Liu, Jian-Ting Ning, Qi Feng et al.· Proceedings of the Thirty-Fi...· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
