Stars
LibriSpeech-Long is a benchmark dataset for long-form speech generation and processing. Released as part of "Long-Form Speech Generation with Spoken Language Models" (arXiv 2024).
Official code for the NeurIPS25 paper "RAT: Bridging RNN Efficiencyand Attention Accuracy in Language Modeling" (https://arxiv.org/abs/2507.04416))
Alleviating Forgetfulness of Linear Attention by Hybrid Sparse Attention and Contextualized Learnable Token Eviction.
A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training
OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.
One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio Tasks
A streamlined and customizable framework for efficient large model (LLM, VLM, AIGC) evaluation and performance benchmarking.
Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparen…
🔥 A minimal training framework for scaling FLA models
OSUM & OSUM-EChat, open speech understanding model and empathetic spoken chatbot based on it, open-sourced by ASLP@NPU.
Joint Fine-tuning and Conversion of Pretrained Speech and Language Models towards Linear Complexity
[ICLR 2025 Oral] NeuralPlane: Structured 3D Reconstruction in Planar Primitives with Neural Fields
Helpful tools and examples for working with flex-attention
FlashInfer: Kernel Library for LLM Serving
[ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule
🚀 Efficient implementations of state-of-the-art linear attention models
Fully open data curation for reasoning models
Official repository of our work "Finding Lottery Tickets in Vision Models via Data-driven Spectral Foresight Pruning" accepted at CVPR 2024
Publication-ready NN-architecture schematics.
The Elements of Statistical Learning (ESL)的中文翻译、代码实现及其习题解答。
The entmax mapping and its loss, a family of sparse softmax alternatives.
Tutorial: Graph Neural Networks for Natural Language Processing at EMNLP 2019 and CODS-COMAD 2020
Summer course on mathematical theory of deep learning
Repo for counting stars and contributing. Press F to pay respect to glorious developers.
MagPhase Vocoder: Speech analysis/synthesis system for TTS and related applications.
A tensorflow based implementation of DeepVoice3 https://arxiv.org/abs/1710.07654