Dahua Lin

The paper introduces AMix-2, a novel protein-text foundation model that unifies protein understanding and sequence design by embedding both modalities in a shared token space, achieving state-of-the-art performance on comprehensive benchmarks.

The paper introduces AMix-2, a novel protein-text foundation model that unifies protein understanding and sequence design by embedding both modalities in a shared token space, achieving state-of-the-art performance on comprehensive benchmarks.

This paper presents SpecGen, a system that optimizes GPU kernel tuning by leveraging speculative generation and dynamic resource allocation.

This paper introduces RNG-Bench, a benchmark suite to evaluate a model's ability to reconstruct past observations and act on them during multi-step interaction.

The paper introduces MemoryWAM, a world action model with efficient persistent memory for robotic manipulation tasks.

The paper presents SenseNova-Vision, a unified multimodal model for computer vision tasks using natural language instructions and optional visual prompts, trained primarily on a new corpus and requiring no task-specific modifications.

This paper introduces the Kinematic-Aware Articulation Interface (KAI), a representation that captures the kinematic structure of articulated objects, improving sample efficiency and generalization in robot manipulation.