SpikingJelly is an open-source deep learning framework for Spiking Neural Network (SNN) based on PyTorch.

An 8-step inversion and 8-step editing process works effectively with the FLUX-dev model. (3x speedup with results that are comparable or even superior to baseline methods)

Learnable Semi-structured Sparsity for Vision Transformers and Diffusion Transformers

This is the official repo of the paper "Latent Guard: a Safety Framework for Text-to-image Generation"

A Survey of Attributions for Large Language Models

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention & RFT)

Memory-Guided Diffusion for Expressive Talking Video Generation

Forcing Diffuse Distributions out of Language Models

Codebase for VidHal: Benchmarking Hallucinations in Vision LLMs

HunyuanVideo: A Systematic Framework For Large Video Generation Model

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

TinyFusion: Diffusion Transformers Learned Shallow

List of papers on Self-Correction of LLMs.

🏠 MEGA: Memory-Efficient 4D Gaussian Splatting for Dynamic Scenes

[NeurIPS 2024] The official implementation of paper: Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs.

Large Reasoning Models

A curated list of Human Preference Datasets for LLM fine-tuning, RLHF, and eval.

CoDe: Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient

Structured Text Generation

Subjects200K dataset

A minimal and universal controller for FLUX.1.

📚 A collection of awesome Causality in ST data papers.

An Open Large Reasoning Model for Real-World Solutions

LLaVA-CoT, a visual language model capable of spontaneous, systematic reasoning

Code to reproduce experiments from the paper Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs