Documents used for grad school application

Official PyTorch Implementation of "SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers"

[NeurIPS'25] KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems

An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.

[NeurIPS'25 Spotlight] Boosting Generative Image Modeling via Joint Image-Feature Synthesis

This is the official Python version of Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Play.

AI Audio Datasets (AI-ADS) 🎵, including Speech, Music, and Sound Effects, which can provide training data for Generative AI, AIGC, AI model training, intelligent audio tool development, and audio a…

Welcome to the FlashSVD, an activation aware inference system for SVD-based low-rank model inference. If you find this repository helpful, please consider starring 🌟 it to support the project — it …

⚡ZEUS accelerates your diffuser. Any modality. Any model. Any scheduler. https://yixiao-wang-stats.github.io/zeus/

Reference PyTorch implementation and models for DINOv3

On Training-Free Acceleration of Generative Modeling

PyTorch implementation of RCG https://arxiv.org/abs/2312.03701

End-to-End Speech Processing Toolkit

ECCV2022 - Real-Time Intermediate Flow Estimation for Video Frame Interpolation

[CVPR 2025 (Oral)] Open implementation of "RandAR"

Official repository for the article Compositional Discrete Latent Code for High Fidelity, Productive Diffusion Models (https://arxiv.org/abs/2507.12318)

[ICLR'25 Spotlight] Min-K%++: Improved baseline for detecting pre-training data of LLMs

[NeurIPS'24] Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy

[ICML 2025] Official Repo for Stability-guided Adaptive Diffusion Acceleration. 🚀🌙Accelerating off-the-shelf diffusion model with a unified stability criterion.

Contrastive Language-Audio Pretraining

[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

A beautiful, simple, clean, and responsive Jekyll theme for academics

🌐 Jekyll is a blog-aware static site generator in Ruby

(SRA) No Other Representation Component Is Needed: Diffusion Transformers Can Provide Representation Guidance by Themselves

(NeurIPS 2024 Oral 🔥) Improved Distribution Matching Distillation for Fast Image Synthesis

Generative models for conditional audio generation

Consistency Models Made Easy

[ICLR 2025] Code for the paper "Tuning Timestep-Distilled Diffusion Model Using Pairwise Sample Optimization"

Implementation of Differentiable Digital Signal Processing (DDSP) in Pytorch