Ranking-Based Explanation Quality Assessment with Listwise Rewards
Discover a novel ranking approach to explanation quality assessment using listwise rewards, improving AI model evaluation and policy optimization.
Credal Concept Bottleneck Models for Uncertainty Decomposition
Explore Credal Concept Bottleneck Models that separate epistemic and aleatoric uncertainty for clearer AI predictions and better decision-making.
Evaluating Sustainable City Trips with LLM and Human Input
Discover a multi-dimensional framework using LLMs and human experts to evaluate sustainable city trips for better travel recommendations.
Right-to-Act: AI Pre-Execution Decision Safety Protocol
Discover the Right-to-Act protocol, a pre-execution AI decision framework ensuring safer, non-compensatory control of AI actions before execution.
Joint vs Modular Learning in Job Shop Scheduling
Explore when joint training outperforms modular learning in job shop scheduling with transportation resources for optimal manufacturing efficiency.