Look Back to Reason Forward- Revisitable Memory for Long-Context LLM Agents
如果无法正常显示，请先停止浏览器的去广告插件。
1. Look Back to Reason Forward: Revisitable Memory
for Long-Context LLM Agents
Yaorui Shi, Yuxin Chen, Siyuan Wang, Sihang Li, Hengxing Cai, Qi Gu, Xiang Wang, An Zhang
USTC · NUS · SJTU · DP Technology · Meituan
Slides by Yaorui Shi
📜 arxiv.org/abs/2509.23040
🤗 huggingface.co/papers/2509.23040
🔗 github.com/syr-cn/ReMemR1
2. Research Background
3. Research Background
4. Motivation
5. Method – Callback Mechanism
6. Method – Callback Mechanism
7. Method – Callback Mechanism
8. Method – Reward Attribution
9. Method – Reward Attribution
10. Method – Reward Attribution
11. Method – Reward Attribution
12. Results
13. Results
14. Results
15. Results
16. Case Study – Baseline (MemAgent)
17. Case Study – ReMemR1
18. Q&A
19. Thank You!
📜 arxiv.org/abs/2509.23040
🤗 huggingface.co/papers/2509.23040
🔗 github.com/syr-cn/ReMemR1