Wei Zhu

The paper proposes COSE, a method that uses an LLM's intrinsic confidence as an uncertainty signal to improve self-evolutionary training, achieving state-of-the-art performance on general reasoning and mathematics.

Qwen-VLA introduces a unified embodied foundation model that extends vision-language understanding to continuous action generation, enabling robust, multi-task generalization across diverse robotic tasks and embodiments.

ReSkill is an RL-in-the-loop framework that reconciles skill creation and policy optimization by automatically creating, testing, and refining modular skills alongside the agent's policy learning, leading to superior generalization.

DFlare introduces a lightweight layer-wise fusion mechanism to overcome the narrow conditioning bottleneck of existing block diffusion methods, enabling the scaling of draft models and achieving superior speculative decoding speedups across multiple LLMs.

The paper introduces LongVQUBench, a comprehensive benchmark for long-term video quality understanding with 1200 diverse videos and 1500 questions.

RecGPT-V3 is a stateful, hybrid-modal recommender system that uses a Memory Hub for user memory and a Hybrid-modal Foundation Model for joint reasoning over text tags and Semantic IDs, achieving consistent gains in user experience and commercial outcomes.

This paper proposes CoSMIC, a framework for semantic integrated sensing and communication (ISAC) that embeds semantic information into waveforms while maintaining covertness and sensing fidelity.