MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent
Authors: Hongli Yu, Tinghong Chen, Jiangtao Feng, Jiangjie Chen, Weinan Dai, Qiying Yu, Ya-Qin Zhang, Wei-Ying Ma, Jingjing Liu, Mingxuan Wang, Hao Zhou
Abstract: Despite improvements by length extrapolation, efficient attention and memory modules, handling infinitely long documents with linear complexity without performance degradation during extrapolation remains the ultimate challenge in long-text processing. We directly optimize for long-text tasks in an end-to-end fashion and introduce a novel agent workflow, MemAgent, which reads text in segments and updates the memory using an overwrite strategy. We extend the DAPO algorithm to facilitate training via independent-context multi-conversation generation. MemAgent has demonstrated superb long-context capabilities, being able to extrapolate from an 8K context trained on 32K text to a 3.5M QA task with performance loss < 5% and achieves 95%+ in 512K RULER test.
Explore the paper tree
Click on the tree nodes to be redirected to a given paper and access their summaries and virtual assistant
Look for similar papers (in beta version)
By clicking on the button above, our algorithm compares the title and abstract of this paper with all the papers of the same arXiv category that we have summarized, and finds the closest ones. You can rerun it from time to time: the results improve as our database grows.