Zijing Hu

NexusAlign. A unified and extensible framework for aligning foundation models.

CIK-Bench. Official repository for Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw

Awesome-RL-for-Video-Generation. A curated list of papers on reinforcement learning for video generation

Awesome-Multimodal-RL. This repository collects research papers on reinforcement learning in generative multimodal models.

ScaleSplat. Scalable SfM-Free 3D Reconstruction with VGGT and Gaussian Splatting

TotalSegmentator. Tool for robust segmentation of >100 important anatomical structures in CT and MR images

Causal_Agent. Causal Agent based on Large Language Model

CAT. Python

AsynDM. [ICLR 26] Asynchronous diffusion models allocate individual pixels with varying timestep schedules, yielding improved text-to-image alignment.

ECIT. Ensemble Conditional Independence Test

DecorrelationDiffusion. [ICCV 2025] Decoding Correlation-Induced Misalignment in the Stable DiffusionWorkflow for Text-to-Image Generation

latent-score-based-reweighting. [ICML 2025] Latent Score-Based Reweighting for Robust Classification on Imbalanced Tabular Data

D-Fusion. [ICML 25] Denoising trajectory fusion, a method to construct RL-trainable visually consistent samples.

STAR-1. [AAAI'26 Oral] Official Implementation of STAR-1: Safer Alignment of Reasoning LLMs with 1K Data

hu-zijing.github.io. Hi, I'm Zijing Hu, a Ph.D student of Zhejiang University.

B2-DiffuRL. [CVPR 25] A framework named B^2-DiffuRL for RL-based diffusion model fine-tuning.