trustworthiness
The degree to which AI systems can be relied upon to produce accurate, fair, and unbiased results, playing a critical role in user adoption and the ethical application of AI technologies.
- BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model
- Do LLMs Really Forget? Evaluating Unlearning with Knowledge Correlation and Confidence Awareness
- Investigating Hallucinations of Time Series Foundation Models through Signal Subspace Analysis
- Learning Human-Like RL Agents Through Trajectory Optimization With Action Quantization
- On the Robustness of Verbal Confidence of LLMs in Adversarial Attacks
- ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation
- The VLLM Safety Paradox: Dual Ease in Jailbreak Attack and Defense