Agents
MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent
arXiv:2507.02259v2 Announce Type: replace Abstract: Despite improvements by length extrapolation, efficient attention and memory modules, handling infinitely long documents with linear complexity with
arXiv:2507.02259v2 Announce Type: replace 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.
Related
- OCR-Memory: Optical Context Retrieval for Long-Horizon Agent Memory
- EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory
- ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents
- MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading
Source: arXiv cs.CL | 2026-07-30