Self-Routing: Parameter-Free Expert Routing for MoE Models
Discover Self-Routing, a parameter-free expert routing method that boosts MoE model efficiency by using hidden states without a learned router.
Decision-Centric Design to Enhance LLM System Reliability
Discover how decision-centric design improves control, reliability, and diagnostics in large language model (LLM) systems for better task success.
Independent Reproduction of OpenAI GPT-OSS-20B Scores
Successfully reproduced OpenAI GPT-OSS-20B scores using reverse-engineered tools and a native harmony agent harness for accurate AI evaluation.
Signal-Based Trajectory Sampling for Efficient Agent Interaction
Optimize agentic interactions with signal-based trajectory sampling and triage for improved efficiency and post-deployment performance.
Collaborative AI for Fault Detection in Network Telemetry
Discover how collaborative AI agents and critics enhance fault detection, cause analysis, and severity assessment in network telemetry systems.