conftrace_

ai safety

235 papers

Explore in graph

Co-occurring keywords

large language model (13587) responsible ai (182) jailbreak attack (225) adversarial attack (1621) language model (4599) adversarial learning (1595) multi-agent system (1819) safety alignment (178) model alignment (226) agent system (324)

Papers

LLM Evaluators Recognize and Favor Their Own Generations NIPS 2024
On scalable oversight with weak LLMs judging strong LLMs NIPS 2024
Do-Not-Answer: Evaluating Safeguards in LLMs EACL 2024
OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research JMLR 2024
Tricking LLMs into Disobedience: Formalizing, Analyzing, and Detecting Jailbreaks COLING 2024
Intersectionality in AI Safety: Using Multilevel Models to Understand Diverse Perceptions of Safety in Conversational AI COLING 2024
Diversity-Aware Annotation for Conversational AI Safety COLING 2024
Implications of Regulations on Large Generative AI Models in the Super-Election Year and the Impact on Disinformation COLING 2024
Flames: Benchmarking Value Alignment of LLMs in Chinese NAACL 2024
Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections NAACL 2024
Enhancing Large Language Models Against Inductive Instructions with Dual-critique Prompting NAACL 2024
XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models NAACL 2024
Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinking NAACL 2024
HalluSafe at SemEval-2024 Task 6: An NLI-based Approach to Make LLMs Safer by Better Detecting Hallucinations and Overgeneration Mistakes NAACL 2024
Universal Prompt Optimizer for Safe Text-to-Image Generation NAACL 2024
Towards Healthy AI: Large Language Models Need Therapists Too NAACL 2024
Discovering Agents (Abstract Reprint) AAAI 2024
How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs ACL 2024
PsySafe: A Comprehensive Framework for Psychological-based Attack, Defense, and Evaluation of Multi-agent System Safety ACL 2024
Compromesso! Italian Many-Shot Jailbreaks undermine the safety of Large Language Models ACL 2024
Sowing the Wind, Reaping the Whirlwind: The Impact of Editing Language Models ACL 2024
Relying on the Unreliable: The Impact of Language Models’ Reluctance to Express Uncertainty ACL 2024
A Comprehensive Study of Jailbreak Attack versus Defense for Large Language Models ACL 2024
Large Language Models Are Involuntary Truth-Tellers: Exploiting Fallacy Failure for Jailbreak Attacks EMNLP 2024
Intrinsic Self-correction for Enhanced Morality: An Analysis of Internal Mechanisms and the Superficial Hypothesis EMNLP 2024