Xi-Yuan Yang

This paper proposes World Model RL (WMRL), which replaces environment execution with a world model to remove this bottleneck and accelerates training by 3-4x on various tasks at different agent scales, while exceeding the performance of standard RL baselines.

Bridging the gap between inference and adaptation in VLMs could lead to significant performance boosts by ensuring robust pseudo-labels that accurately reflect sample-level relationships.

Progress reward modeling could revolutionize how robots learn in dynamic environments by providing nuanced feedback beyond mere task completion.

LLM agents struggle significantly with personalized tool use, revealing critical gaps in their capabilities that existing benchmarks overlook.

Domain-specific scientific models, previously siloed from LLM agent systems, can now be orchestrated for complex reasoning tasks via the Eywa framework, unlocking performance gains on structured data.

Looping language models isn't just for single agents anymore: Recursive Multi-Agent Systems (RecursiveMAS) show that agent collaboration itself can be scaled through recursion, yielding faster and more efficient problem-solving.

Ramen achieves robust test-time adaptation of VLMs in mixed-domain scenarios by selecting the right samples to adapt to, sidestepping the common pitfall of performance degradation when faced with diverse and inconsistent test data.