Jie Zhou

MathForm: Scaling Mathematical Autoformalization with Knowledge Retrieval and Verification-Guided Refinement

Lushi Pu, Weiming Zhang, Xinheng Xie, et al.

AgentHPOBench: A Benchmark For Evaluating LLM Agents as Sequential Hyperparameter Optimizers

Tianyu Huai, Tingshuo Fan, Xinchi Chen, et al.

SM4RT: Learning Structured Motion Geometry for 4D Reconstruction

Shing Ho J. Lin, Wenzhao Zheng, Dong Zhuo, et al.

UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing

Xinlong Zhao, Dongsheng Liu, Hengyu Zhao, et al.

Agents-K1: Towards Agent-native Knowledge Orchestration

Zongsheng Cao, Bihao Zhan, Jinxin Shi, et al.

BAMI: Training-Free Bias Mitigation in GUI Grounding

Borui Zhang, Bo Zhang, Bo Wang, et al.

Contextual Multi-Objective Optimization: Rethinking Objectives in Frontier AI Systems

Jie Zhou, Qin Chen, Liang He

Action-Aware Generative Sequence Modeling for Short Video Recommendation

Wenhao Li, Zihan Lin, Zhengxiao Guo, et al.

Ask Only When Needed: Proactive Retrieval from Memory and Skills for Experience-Driven Lifelong Agents

Yuxuan Cai, Jie Zhou, Qin Chen, et al.

ChatSVA: Bridging SVA Generation for Hardware Verification via Task-Specific LLMs

Lik Tung Fu, Jie Zhou, Shaokai Ren, et al.

Attention at Rest Stays at Rest: Breaking Visual Inertia for Cognitive Hallucination Mitigation

Boyang Gong, Yu Zheng, Fanye Kong, et al.

PsychAgent: An Experience-Driven Lifelong Learning Agent for Self-Evolving Psychological Counselor

Yutao Yang, Junsong Li, Qianjun Pan, et al.

Vega: Learning to Drive with Natural Language Instructions

Sicheng Zuo, Yuxuan Li, Wenzhao Zheng, et al.

DriveTok: 3D Driving Scene Tokenization for Unified Multi-View Reconstruction and Understanding

Dong Zhuo, Wenzhao Zheng, Sicheng Zuo, et al.

DriveTok is a new tool that converts multi-camera driving scenes into efficient digital 'tokens' (compressed representations) that capture semantic meaning, depth, and 3D spatial information all at once. Unlike existing tokenizers designed for single images, DriveTok is specifically built for autonomous vehicles with multiple cameras, using advanced attention mechanisms to ensure consistency across different camera views. The tokens can be used for various driving-related tasks like reconstructing images, segmenting objects, predicting depth, and understanding 3D space around the vehicle.