SenseNova-U series: Native Unified Paradigm with NEO-unify from the First Principles

LLaDA2.0-Uni: Understanding and Generation the World.

Job Talks @ Mines

DVD: Deterministic Video Depth Estimation with Generative Priors

Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

scDFM: Distributional Flow Matching for Robust Single-Cell Perturbation Prediction (ICLR 2026)

[ICML 2026] LatentMorph: Morphing Latent Reasoning into Image Generation

A UX system for full scale deployment of a llm driven video editing ysstem

[ICLR26 Oral] RealPDEBench: A Benchmark for Complex Physical Systems with Paired Real-World and Simulated Data

Official implementation of “4D LangVGGT: 4D Language-Visual Geometry Grounded Transformer”

[CVPR 2026] TiViBench: Benchmarking Think-in-Video Reasoning for Video Generative Models

BuildArena, where LLM agents design, build, and test rockets, cars, and bridges in a physics simulator given a goal-directed sentence.

[ICML 2026] ScalingAR: Scaling Confidence for Autoregressive Image Generation

Explore how to get a VQ-VAE models efficiently!

An Efficient Text-to-Image Generation Pretrain Pipeline

[🚀 ICLR 2026 Oral] NextStep-1: SOTA Autogressive Image Generation with Continuous Tokens. A research project developed by the StepFun’s Multimodal Intelligence team.

Reference PyTorch implementation and models for DINOv3

TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and Generation

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

[ICCV 2025] FiVE-Bench: A Fine-grained Video Editing Benchmark for Evaluating Emerging Diffusion and Rectified Flow Models

A curated list of awesome papers for reconstructing 4D spatial intelligence from video. (arXiv 2507.21045)

Wan: Open and Advanced Large-Scale Video Generative Models

[CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

[ICLR 2026] Streaming 4D Visual Geometry Transformer