From 5b1d12a29f76b7f1d68083f93a33f9ddb78b0af2 Mon Sep 17 00:00:00 2001 From: ylr9933 <2835703479@qq.com> Date: Sat, 11 Jul 2026 20:26:37 +0800 Subject: [PATCH 1/2] reorganize related projects section, add full LightMem series links --- README.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/README.md b/README.md index 0a71391..56f875e 100644 --- a/README.md +++ b/README.md @@ -642,6 +642,13 @@ We welcome bug fixes, host adapter improvements, onboarding fixes, tests, and do We thank all the contributors to this project, more contributors are welcome! + +### πŸ“š Our LightMem Series +This repository belongs to ZJU NLP LightMem lightweight agent memory series, focusing on solving context bloat, excessive token consumption and low cache utilization for long-running LLM agents: +- [LightMem](https://github.com/zjunlp/LightMem.git) β€” ICLR 2026 lightweight memory-augmented generation baseline, core hierarchical memory paradigm +- [LightMem2](https://github.com/zjunlp/LightMem2) β€” This repo, multi-host runtime framework with TokenPilot for code agents +- [LightMem-Ego](https://github.com/zjunlp/LightMem-Ego) β€” Temporal memory extension for embodied & ego-centric autonomous agents + #### Other Related Projects - [LLMLingua-2](https://github.com/microsoft/LLMLingua) β€” Token-level prompt compression From c683363136edd9880d7452babef5fa23abb22131 Mon Sep 17 00:00:00 2001 From: ylr9933 <2835703479@qq.com> Date: Sat, 11 Jul 2026 20:58:24 +0800 Subject: [PATCH 2/2] update readme. --- README.md | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 56f875e..47e3f81 100644 --- a/README.md +++ b/README.md @@ -67,6 +67,7 @@ LightMem2 is intended to host multiple long-running-agent components over time. * πŸ“„ Citation * 🀝 Contributing * πŸŽ‰ Contributors +* πŸ“š Related Works * πŸ’¬ Community @@ -634,6 +635,7 @@ Please cite our paper if you use LightMem2 in your work. We welcome bug fixes, host adapter improvements, onboarding fixes, tests, and documentation updates, see [CONTRIBUTING.md](./CONTRIBUTING.md) for more details. + ## πŸŽ‰Contributors @@ -643,13 +645,17 @@ We welcome bug fixes, host adapter improvements, onboarding fixes, tests, and do We thank all the contributors to this project, more contributors are welcome! -### πŸ“š Our LightMem Series -This repository belongs to ZJU NLP LightMem lightweight agent memory series, focusing on solving context bloat, excessive token consumption and low cache utilization for long-running LLM agents: -- [LightMem](https://github.com/zjunlp/LightMem.git) β€” ICLR 2026 lightweight memory-augmented generation baseline, core hierarchical memory paradigm -- [LightMem2](https://github.com/zjunlp/LightMem2) β€” This repo, multi-host runtime framework with TokenPilot for code agents -- [LightMem-Ego](https://github.com/zjunlp/LightMem-Ego) β€” Temporal memory extension for embodied & ego-centric autonomous agents + -#### Other Related Projects +## πŸ“š Related Works + +### LightMem Series +This repository belongs to ZJUNLP LightMem series, focusing on solving context bloat, excessive token consumption and low cache utilization for long-running LLM agents: +- [LightMem](https://github.com/zjunlp/LightMem) β€” Lightweight memory-augmented generation +- [LightMem2](https://github.com/zjunlp/LightMem2) β€” This repo, multi-host runtime modular framework for agents +- [LightMem-Ego](https://github.com/zjunlp/LightMem-Ego) β€” Temporal memory extension for ego-centric autonomous agents + +### Other Related Projects - [LLMLingua-2](https://github.com/microsoft/LLMLingua) β€” Token-level prompt compression - [SelectiveContext](https://github.com/liyucheng09/Selective_Context) β€” Self-information-based context reduction @@ -657,7 +663,6 @@ This repository belongs to ZJU NLP LightMem lightweight agent memory series, foc - [MemoBrain](https://github.com/qhjqhj00/MemoBrain) β€” Executive memory for long-horizon reasoning agents - [AgentSwing](https://github.com/Alibaba-NLP/DeepResearch) β€” Adaptive parallel context management routing for web agents - [MemOS](https://github.com/MemTensor/MemOS) β€” Memory operating system for LLM agents -- [LightMem](https://github.com/zjunlp/LightMem) β€” Lightweight memory-augmented generation - [Headroom](https://github.com/chopratejas/headroom) β€” Compresses everything when AI agent reads