Wenhao Li

GeoProp achieves a remarkable 10.6% boost in real-world manipulation tasks by effectively grounding robot state in visual context, all while adding minimal complexity.

EP-SAM outperforms traditional SAM methods by effectively addressing contour ambiguity in ultrasound images through innovative edge-aware supervision.

Robots can now autonomously adapt to camera changes without needing explicit calibration, significantly improving deployment flexibility.

RaBitQCache accelerates long-context LLM inference while cutting memory I/O by intelligently adapting token budgets based on attention sparsity.

AgentX can autonomously iterate on recommendation algorithms, outpacing human-driven processes and fundamentally changing how we approach system development.

Real-robot trials are costly and slow, but DataLadder enables scalable evaluations and data generation through a seamless interplay between robots, simulations, and human feedback.

LLMs struggle to accurately evaluate real human reasoning in mathematics, revealing a critical evaluation gap that challenges current assessment methods.

FlashTTS slashes First-Packet Latency to 325ms, revolutionizing real-time speech dialogue systems without sacrificing voice quality.

PEFT can enable the creation of millions of personalized models, each with unique adaptations, leveraging the power of trillion-parameter foundation models.

LLM-based multi-agent systems can be optimized far more efficiently by decomposing credit assignment temporally and structurally, pinpointing weak links for targeted refinement.

Scaling offline MARL to thousands of agents is now tractable: MF-Diffuser uses mean-field theory to plan in trajectory distribution space, sidestepping the curse of dimensionality.

AgentSchool offers a powerful new way to simulate educational environments, moving beyond simple role-play to model learning as a dynamic state transition and providing a testbed for long-horizon memory and multi-agent coordination.