Popular GitHub Repositories for Length Extrapolation

MuseV: Infinite-length and High Fidelity Virtual Human Video Generation with Visual Conditioned Parallel Denoising

? Must-read papers and blogs on LLM based Long Context Modeling ?

AKGR: Awesome Knowledge Graph Reasoning is a collection of knowledge graph reasoning works, including papers, codes and datasets

Image Outpainting and Harmonization using GANs - PyTorch Implementation (official)

High-fidelity performance metrics for generative models in PyTorch

Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The purpose of this tool is to solve the massive Transpose extrapolation problem in onnx-tensorflow (onnx-tf). I don't need a Star, but give me a pull request.

Transformer based on a variant of attention that is linear complexity in respect to sequence length

Implementation of Phenaki Video, which uses Mask GIT to produce text guided videos of up to 2 minutes in length, in Pytorch

Implementation of plug in and play Attention from "LongNet: Scaling Transformers to 1,000,000,000 Tokens"

Official implementation for "RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers" (ICML 2025)

Extend existing LLMs way beyond the original training length with constant memory usage, without retraining

[NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization

Natural language processing pipeline for book-length documents (archival Java version; for current Python version, see: https://github.com/booknlp/booknlp)

AIxplora is a open-source tool which let's you query all kind of files not limited to any length or format.

ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation (CVPR20)

[CVPR 2024] Official Implementation of "Seamless Human Motion Composition with Blended Positional Encodings".

One-Stop Solution to encode sentence to fixed length vectors from various embedding techniques

?ReVisual-R1 is a 7B open-source multimodal language model that follows a three-stage curriculum—cold-start pre-training, multimodal reinforcement learning, and text-only reinforcement learning—to achieve faithful, concise, and self-reflective state-of-the-art performance in visual and textual reasoning.

LongQLoRA: Extent Context Length of LLMs Efficiently

Implementation of NAACL 2024 Outstanding Paper "LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models"