Enhancing AI Reasoning with Executor-Grounded Rewards
Discover how executor-grounded rewards improve AI reasoning quality beyond correctness using the TraceLift framework and new TRACELIFT-GROUPS dataset.
Mechanical Conscience: Ensuring Dependable Machine Intelligence
Discover Mechanical Conscience, a novel framework enhancing safety and dependability in collaborative machine intelligence under uncertainty.
SOAR: Real-Time Optimization for Robot Scheduling & Orders
Discover SOAR, a real-time framework optimizing order allocation and robot scheduling in robotic mobile fulfillment systems for enhanced warehouse efficien...
Agentic-imodels: Advancing Autonomous Data Science Tools
Discover Agentic-imodels, evolving autonomous data science tools with LLM-based interpretability and improved predictive performance for AI agents.
ScrapMem: Efficient On-Device Memory for AI Agents
Discover ScrapMem, a bio-inspired framework that boosts AI agent memory with optical forgetting, cutting storage by 93% and improving recall.