Official implementation of our NeurIPS 2023 paper "Augmenting Language Models with Long-Term Memory".
-
Updated
Mar 30, 2024 - Python
Official implementation of our NeurIPS 2023 paper "Augmenting Language Models with Long-Term Memory".
[EMNLP 2024 (Oral)] Leave No Document Behind: Benchmarking Long-Context LLMs with Extended Multi-Doc QA
LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models
[ICLR 2024] CLEX: Continuous Length Extrapolation for Large Language Models
🔥 [ICLR 2025] Official Benchmark Toolkits for "Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark"
This repo contains evaluation code for the paper "MileBench: Benchmarking MLLMs in Long Context"
[ICLR 2025] Official code repository for "TULIP: Token-length Upgraded CLIP"
[NeurIPS 2025] HoPE: Hybrid of Position Embedding for Long Context Vision-Language Models
🔥 [ICLR 2025] Official PyTorch Model "Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark"
Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers
context denoising training for long-context modeling
🫧 Code for Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data (Maekawa*, Iso* et al.; ICLR 2025)
CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs
HiCI: Hierarchical Construction-Integration for Long-Context Attention
Pure PyTorch + 🤗 Transformers reimplementation of Megalodon (CEMA + chunked attention) - readable, hackable, no CUDA kernels required
Minimal implementation of Samba by Microsoft in PyTorch
This is the official implementations for SustainableKV
jax rewrite for CEMA support + other training speedups.
To associate your repository with the long-context-modeling topic, visit your repo's landing page and select "manage topics."