Replace-then-Perturb: Targeted Adversarial Attacks With Visual Reasoning for Vision-Language Modelsby Jonggyu Jang, Hyeonsu Lyu, Jungyeon…
Schema Augmentation for Zero-Shot Domain Adaptation in Dialogue State Trackingby Christopher Richardson, Roshan Sharma, Neeraj…
Generating Diverse Negations from Affirmative Sentencesby Darian Rodriguez Vasquez, Afroditi PapadakiFirst submitted to arxiv on:…
Evolving Alignment via Asymmetric Self-Playby Ziyu Ye, Rishabh Agarwal, Tianqi Liu, Rishabh Joshi, Sarmishta Velury,…
Beyond Ontology in Dialogue State Tracking for Goal-Oriented Chatbotby Sejin Lee, Dongha Kim, Min SongFirst…
InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Modelsby Hao Li, Xiaogeng LiuFirst submitted…
Stealthy Jailbreak Attacks on Large Language Models via Benign Data Mirroringby Honglin Mu, Han He,…
Palisade – Prompt Injection Detection Frameworkby Sahasra Kokkula, Somanathan R, Nandavardhan R, Aashishkumar, G DivyaFirst…
LLM Robustness Against Misinformation in Biomedical Question Answeringby Alexander Bondarenko, Adrian ViehwegerFirst submitted to arxiv…
Fine-tuned Large Language Models (LLMs): Improved Prompt Injection Attacks Detectionby Md Abdur Rahman, Fan Wu,…