TSLANet

lpwm 开源

用潜在粒子建模视频世界，预测未来更精准

[ICLR 2026 Oral] Latent Particle World Models official repository

runbooks 开源

在K8s上声明式微调LLM，一键跑通训练全流程

Finetune LLMs on K8s by using Runbooks

LightRFT 开源

用奖励模型驱动微调，让 LLM/VLM 按目标越训越强

LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuni···

gflownet 开源

按奖励比例采样多样化候选解的生成模型框架

Generative Flow Networks - GFlowNet

Matlab-GAN 开源

用 MATLAB 跑通多种 GAN，从入门到 CycleGAN 一步到位

MATLAB implementations of Generative Adversarial Networks -- from GAN to Pixel2Pixel···

opacus 开源

几行代码给 PyTorch 训练加差分隐私，防数据泄露

Training PyTorch models with differential privacy

llm-mcts 开源

用蒙特卡洛树搜索增强LLM，让日常任务规划更靠谱

[NeurIPS 2023] We use large language models as commonsense world model and heuristic···

ai8x-training 开源

在电脑上训练量化模型，一键部署到毫瓦级 AI 芯片

Model Training for ADI's MAX78000 and MAX78002 Edge AI Devices

model-optimization 开源

几行代码压缩模型，让大模型跑进手机和嵌入式设备

A toolkit to optimize ML models for deployment for Keras and TensorFlow, including q···