10 posts published in the last hour
- 04:32Cross-Disciplinary Taxonomy and Modeling of Misunderstanding Generation, Amplification, and Detection, from Pragmatics to AI Agents
- 04:32AI Evaluation Should Work With Humans
- 04:32Stable Miscalibration in Large Language Models: A Practical View of High-Confidence Errors
- 04:32Active Perception for Embodied Disambiguation
- 04:31Measuring Cross-Task Behavioral Consistency in Language Model Agents
- 04:03Modular Cognitive Architecture Emerges in Large Language Models
- 04:03Inducing Reward-Free Judging Rubrics that Reduce Over-Crediting in Agent Evaluation
- 04:03Agentao: A Governed Local-First Runtime for Tool-Using LLM Agents
- 04:03A Year in LLM Serving: Workload Evolution, Caching and Load-Balancing
- 04:03Depth-Aware Sensitivity Analysis of Mixture-of-Experts Models via Magnitude-Based Expert Masking
