EPO-Safe: Learning AI Safety from 1-Bit Danger Signals
EPO-Safe enables AI agents to discover safety rules using minimal 1-bit danger signals, enhancing autonomous safe behavior in complex environments.
StoryTR: Video Retrieval with Theory of Mind Reasoning
Discover StoryTR, a benchmark for narrative-centric video retrieval using Theory of Mind to enhance understanding of character intent and story context.
Self-Adaptive Hierarchical Planning for Efficient LLM Agents
Discover AdaPlan-H, a self-adaptive hierarchical planning method that improves LLM agents' task execution by balancing coarse and fine planning.
Bias Mitigation in LLM Judges: Effective Strategies Tested
Explore key bias types in LLM judge models and discover the most effective strategies to reduce bias in AI evaluation pipelines.
PhySE: Real-Time AR-LLM Social Engineering Framework
Discover PhySE, a cutting-edge framework addressing real-time AR and LLM social engineering attacks with adaptive psychological tactics.