zhaoyl18 / SEIKO
SEIKO is a novel reinforcement learning method to efficiently fine-tune diffusion models in an online setting. Our methods outperform all baselines (PPO, classifier-based guidance, direct reward backpropagation) for fine-tuning Stable Diffusion.
☆19Updated 7 months ago
Alternatives and similar repositories for SEIKO:
Users that are interested in SEIKO are comparing it to the libraries listed below
- Derivative-Free Guidance in Diffusion Models with Soft Value-Based Decoding. For controlled generation in DNA, RNA, proteins, molecules (…☆28Updated 5 months ago
- Code for the tutorial/review paper for RL-based-fine-tuniing. In this code, we especially focus on the design of biological sequences li…☆111Updated 5 months ago
- Official Code for Local Search GFlowNets (ICLR 2024 Spotlight)☆16Updated 2 weeks ago
- Official codebase for Exact Energy-Guided Diffusion Sampling via Contrastive Energy Prediction☆26Updated last year
- Code for paper: "Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Design"☆36Updated last week
- Code for the paper https://arxiv.org/abs/2402.04997☆70Updated last year
- Implementation of Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Models☆67Updated last year
- This repository is the official implementation of Bidirectional Learning for Offline Infinite-width Model-based Optimization (NeurIPS 202…☆15Updated 2 years ago
- ☆15Updated last year
- ☆24Updated 11 months ago
- ☆31Updated 8 months ago
- Official Code for Paper "Think While You Generate: Discrete Diffusion with Planned Denoising" [ICLR 2025]☆50Updated 3 weeks ago
- Simple Guidance Mechanisms for Discrete Diffusion Models☆26Updated 2 months ago
- Graph Diffusion Policy Optimization☆32Updated 11 months ago
- Code for the paper Iterated Denoising Energy Matching for Sampling from Boltzmann Densities.☆47Updated 3 weeks ago
- Code repository for Trajectory Flow Matching☆56Updated 4 months ago
- Official implementation of ICML'24 paper "Offline Multi-Objective Optimization".☆17Updated 4 months ago
- Baselines for Model-Based Optimization installation fixes and compatible with newer AMPERE+ GPUs (e.g. 3090)☆11Updated last year
- ☆12Updated last year
- A general framework for inference-time scaling and steering of diffusion models with arbitrary rewards.☆103Updated 3 weeks ago
- ☆24Updated last year
- ☆24Updated 5 months ago
- ☆23Updated 7 months ago
- ☆14Updated 4 months ago
- Educational implementation of the Discrete Flow Matching paper☆78Updated 6 months ago
- Entropic Optimal Transport Benchmark (NeurIPS 2023).☆20Updated 11 months ago
- ☆16Updated 11 months ago
- Benchmarks for Model-Based Optimization☆88Updated 10 months ago