Highlights
Lists (1)
Stars
[ICLR 2025 Spotlight] An open-sourced LLM judge for evaluating LLM-generated answers.
A yabai and AeroSpace status bar widget for Übersicht
Code for "SED-SFT: Selectively Encouraging Diversity in Supervised Fine-Tuning"
QZone History 是一个旨在还原和重建 QQ 空间历史的项目。通过收集和分析用户的活动记录，即使某些说说、留言板消息等内容已被删除，本项目也能够进行推理并进行数据重建，帮助用户找回珍贵的历史回忆。
Nix - Static, Permissionless, Installation-free, Pre-configured
Claude Code skills for running research code on HPC clusters
SkyRL: A Modular Full-stack RL Library for LLMs
Pytorch Library for Relational Table Learning with LLMs.
[ICLR 2026] LLM/VLM gaming agents and model evaluation through games.
Dr. MAS is an end-to-end RL training framework for multi-agent LLM systems, supporting the co-training of multiple (heterogeneous) LLMs.
Code for "Sample Efficient Demonstration Selection for In-Context Learning" accepted at ICML 2025
Do Influence Functions Work on Large Language Models?
The open source developer platform to build AI agents and models with confidence. Enhance your AI applications with end-to-end tracking, observability, and evaluations, all in one integrated platform.
An Obsidian plugin that embeds Claude Code as an AI collaborator in your vault
Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.
Official implementation of Coresets from Trajectories: Selecting Data via Correlation of Loss Differences
Differentiable architecture search for convolutional and recurrent networks
Claude Code for CUDA. Free AI assistant that actually understands GPU architecture
NASBench: A Neural Architecture Search Dataset and Benchmark
Pixel perfect icon theme for Linux
[DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"
A foundation model for universal 3D blood vessel segmentation. Paper accepted @ CVPR25.
A large-scale dataset of both raw MRI measurements and clinical MRI images.