Biqing Qi

ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data

Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles

OS-MAP: How Far Can Computer-Using Agents Go in Breadth and Depth?

MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and Inference

LATTS: LAtent space Test Time Scaling for diffusion language models

ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows

SpectrumWorld: Artificial Intelligence Foundation for Spectroscopy

Sequential Diffusion Language Models

Self Speculative Decoding for Diffusion Large Language Model

Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism

OS-Catalyst: Advancing Computer-Using Agents Efficiency through Adaptive Action Compression

MolSpectLLM: A Molecular Foundation Model Bridging Spectroscopy, Molecule Elucidation, and 3D Structure Generation

CLiMRS: Cooperative Large-Language-Model-Driven Heterogeneous Multi-Robot System

Bohdi: Heterogeneous LLM Fusion with Automatic Data Exploration

TTRL: Test-Time Reinforcement Learning

OpenPRM: Building Open-domain Process-based Reward Models with Preference Trees

DePass: Unified Feature Attributing by Simple Decomposed Forward Pass

Fourier Position Embedding: Enhancing Attention’s Periodic Extension for Length Generalization

Fast and Slow Generating: An Empirical Study on Large and Small Language Models Collaborative Decoding