Cost-Efficient Explainable AI Using Uncertainty Gating
Enhance explainable AI reliability and reduce costs by leveraging epistemic uncertainty to optimize explanation generation and resource allocation.
C-TRAIL: Reliable Commonsense Trajectory Planning for AVs
C-TRAIL enhances autonomous driving by integrating commonsense reasoning with trust mechanisms for safer, reliable trajectory planning.
ATP-Bench: Benchmark for Agentic Tool Planning in MLLMs
Discover ATP-Bench, a benchmark for agentic tool planning in multimodal large language models enabling coherent interleaved text-image generation.
Information-Theoretic Model for Rational Categorization
Explore a new information-theoretic approach to rational categorization that advances cognitive science and explains classic experimental findings.
ShapE-GRPO: Improved Reward Allocation for Multi-Candidate LLMs
Discover ShapE-GRPO, a Shapley value-based method enhancing reward allocation for multi-candidate LLM training, boosting performance and exploration.