Elamur Research⌕
☰
home / papers / 2026 / CLAG: Adaptive Memory Organization via Agent-Driven Clustering for Small Language Model Agents

CLAG: Adaptive Memory Organization via Agent-Driven Clustering for Small Language Model Agents

Paper · 2026 · Taeyun Roh, WonJune Jang, Junha Jung et al.

Summary

Memory-augmented agents usually keep every stored experience in one flat retrieval pool. As that pool grows, retrieval surfaces more plausible but off-topic entries, and memory-updating ("evolution") steps operate on mixed-topic neighbourhoods that can gradually corrupt the store. Small language models (SLMs, under ~2B parameters) are especially sensitive to such distracting context. CLAG addresses this by letting the SLM agent itself organise memory into clusters.

  • Writing memories: each new memory note is routed to a cluster. Vector distance first shortlists candidate clusters, then the SLM picks among them using cluster profiles (summaries and tags). Clusters that grow too large are split.
  • Updating memories: evolution (linking, rewriting) happens only inside the chosen cluster.
  • Answering queries: a two-stage retrieval first selects relevant clusters from their profiles, then searches only within those clusters.

Across QA benchmarks (LoCoMo, HotpotQA, BioASQ) with three SLM backbones, CLAG improves answer quality and robustness over earlier agent-memory systems while staying lightweight. This makes it relevant for on-device and low-cost agents.

Key points

  • Architecture: a single backbone SLM handles routing, evolution and cluster selection through role-specific prompts, with no separately trained modules. Notes follow an A-mem-style schema (content, cluster, timestamp, keywords, tags, context, links). Clusters are created after a cold-start buffer and are bisected with K-Means once they pass a size threshold.
  • Baselines: compared against RAG, A-mem, GAM and MemoryOS.
  • Agentic vs. geometric clustering on BioASQ: CLAG reaches F1 22.01 and BLEU-1 17.23. The best non-agentic option, K-Means clustering, reaches 15.64 and 13.36. CLAG's evidence-retrieval E-F1 on BioASQ is 25.11.
  • Scaling to Qwen3-8B:
    • CLAG has the best F1 on LoCoMo (21.50).
    • It is best on both HotpotQA metrics (25.40 F1 / 23.53 BLEU-1).
    • On BioASQ it has the best BLEU-1 (23.54), but MemoryOS has the higher F1 (28.83).
    • CLAG's advantage shrinks as model size increases.
  • Limitations and availability: results depend on prompt design and model version, and robustness to distribution shift was not tested. Code is available at github.com/dmis-lab/CLAG.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Korea UniversitySeoul · South Korea26 works · 9,490 citations🎓Myongji UniversitySeoul · South Korea2 works

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksInformation Systems · 1,437 works

Cite

Cite as Taeyun Roh, WonJune Jang, Junha Jung and Jaewoo Kang. 2026. CLAG: Adaptive Memory Organization via Agent-Driven Clustering for Small Language Model Agents. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.824

@inproceedings{roh2026clag,
  title = {{CLAG: Adaptive Memory Organization via Agent-Driven Clustering for Small Language Model Agents}},
  author = {Taeyun Roh, WonJune Jang, Junha Jung and Jaewoo Kang},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.824},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.824}
}

References

43 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Satanjeev Banerjee and Alon Lavie. 2005. Meteor: An automatic metric for mt evaluation with improved correlation with human judgments. In Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization.
  2. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. 2025. From local to global: A graph rag approach to query-focused summarization. Preprint, arXiv:2404.16130. find
  3. Asaf Gilboa and Hannah Marlatte. 2017. Neurobiology of schemas and schema-mediated memory. Trends in cognitive sciences, 21(8):618–631. find
  4. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  5. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948.
  6. Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang, Boyang Liu, Fangyi Zhu, Jiahang Lin, Honglin Guo, Shihan Dou, Zhiheng Xi, Senjie Jin, Jiejun Tan, Yanbin Yin, Jiongnan Liu, Zeyu Zhang, Zhongxiang Sun, Yutao Zhu, Hao Sun, Boci Peng, and 28 others. 2025. Memory in the age of ai agents. Preprint, arXi… — [arXiv:2512.13564](https://arxiv.org/abs/2512.13564) find
  7. Kalervo Järvelin and Jaana Kekäläinen. 2002. Cumulated gain-based evaluation of ir techniques. In ACM Transactions on Information Systems, pages 422– 446. find
  8. Jiazheng Kang, Mingming Ji, Zhe Zhao, and Ting Bai. 2025. Memory os of ai agent. arXiv preprint arXiv:2506.06326.
  9. Dharshan Kumaran, Demis Hassabis, and James L Mc-Clelland. 2016. What learning systems do intelligent agents need? complementary learning systems theory updated. Trends in cognitive sciences, 20(7):512– 534. find
  10. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, and 1 others. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information processing systems, 33:9459–…
  11. Guohao Li, Hasan Hammoud, Hani Itani, Dmitrii Khizbullin, and Bernard Ghanem. 2023. Camel: Communicative agents for" mind" exploration of large language model society. Advances in Neural Information Processing Systems, 36:51991–52008.
  12. Zhenyan Lu, Xiang Li, Dongqi Cai, Rongjie Yi, Fangming Liu, Xiwen Zhang, Nicholas D Lane, and Mengwei Xu. 2024. Small language models: Survey, measurements, and insights. arXiv preprint arXiv:2409.15790.
  13. Xinbei Ma, Yeyun Gong, Pengcheng He, Hai Zhao, and Nan Duan. 2023. Query rewriting in retrieval-augmented large language models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 5303–5315.
  14. Adyasha Maharana, Dong-Ho Lee, Sergey Tulyakov, Mohit Bansal, Francesco Barbieri, and Yuwei Fang. 2024. Evaluating very long-term conversational memory of llm agents. Preprint, arXiv:2402.17753. find
  15. Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, and Hannaneh Hajishirzi. 2023. When not to trust language models: Investigating effectiveness of parametric and non-parametric memories. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (V… find
  16. Charles Packer, Sarah Wooders, Kevin Lin, Vivian Fang, Shishir G. Patil, Ion Stoica, and Joseph E. Gonzalez. 2024. Memgpt: Towards llms as operating systems. Preprint, arXiv:2310.08560. find
  17. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL), pages 311–318.
  18. Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. 2023. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th annual acm symposium on user interface software and technology, pages 1–22. find
  19. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. Squad: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP).
  20. Alireza Rezazadeh, Zichao Li, Wei Wei, and Yujia Bao. 2024. From isolated conversations to hierarchical schemas: Dynamic tree memory representation for llms. arXiv preprint arXiv:2410.14052. — [arXiv:2410.14052](https://arxiv.org/abs/2410.14052) find
  21. Parth Sarthi, Salman Abdullah, Aditi Tuli, Shubh Khanna, Anna Goldie, and Christopher D. Manning. 2024. Raptor: Recursive abstractive processing for tree-organized retrieval. Preprint, arXiv:2401.18059.
  22. Freda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales, David Dohan, Ed H Chi, Nathanael Schärli, and Denny Zhou. 2023. Large language models can be easily distracted by irrelevant context. In International Conference on Machine Learning, pages 31210–31227. PMLR. rise and potential of large language…
  23. Noah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. 2023. Reflexion: Language agents with verbal reinforcement learning. Advances in Neural Information Processing Systems, 36:8634–8652.
  24. Wujiang Xu, Zujie Liang, Kai Mei, Hang Gao, Juntao Tan, and Yongfeng Zhang. 2025. A-mem: Agentic memory for llm agents. Preprint, arXiv:2502.12110.
  25. George Tsatsaronis, Georgios Balikas, Prodromos Malakasiotis, Ioannis Partalas, Matthias Zschunke, Michael R Alvers, Dirk Weissenborn, Anastasia Krithara, Sergios Petridis, Dimitris Polychronopoulos, and 1 others. 2015. An overview of the bioasq large-scale biomedical semantic indexing and question… find
  26. Dorothy Tse, Rosamund F Langston, Masaki Kakeyama, Ingrid Bethus, Patrick A Spooner, Emma R Wood, Menno P Witter, and Richard GM Morris. 2007. Schemas and memory consolidation. Science, 316(5821):76–82. find
  27. Bing Wang, Xinnian Liang, Jian Yang, Hui Huang, Shuangzhi Wu, Peihao Wu, Lu Lu, Zejun Ma, and Zhoujun Li. 2025a. Scm: Enhancing large language model with self-controlled memory framework. Preprint, arXiv:2304.13343. — [arXiv:2304.13343](https://arxiv.org/abs/2304.13343) find
  28. Guanzhi Wang, Yuqi Xie, Yunfan Jiang, Ajay Mandlekar, Chaowei Xiao, Yuke Zhu, Linxi Fan, and Anima Anandkumar. 2023a. Voyager: An open-ended embodied agent with large language models. Preprint, arXiv:2305.16291. find
  29. Lei Wang, Chen Ma, Xueyang Feng, Zeyu Zhang, Hao Yang, Jingsen Zhang, Zhiyuan Chen, Jiakai Tang, Xu Chen, Yankai Lin, and 1 others. 2024. A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6):186345.
  30. Liang Wang, Nan Yang, and Furu Wei. 2023b. Query2doc: Query expansion with large language models. arXiv preprint arXiv:2303.07678.
  31. Zilong Wang, Zifeng Wang, Long Le, Huaixiu Steven Zheng, Swaroop Mishra, Vincent Perot, Yuwei Zhang, Anush Mattapalli, Ankur Taly, Jingbo Shang, Chen-Yu Lee, and Tomas Pfister. 2025b. Speculative rag: Enhancing retrieval augmented generation through drafting. Preprint, arXiv:2407.08223. — [arXiv:2407.08223](https://arxiv.org/abs/2407.08223) find
  32. Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, Rui Zheng, Xiaoran Fan, Xiao Wang, Limao Xiong, Yuhao Zhou, Weiran Wang, Changhao Jiang, Yicheng Zou, Xiangyang Liu, and 10 others. 2023. The rise and potential of large language mode…
  33. Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, and 1 others. 2025. find
  34. The BY Yan, Chaofan Li, Hongjin Qian, Shuqi Lu, and Zheng Liu. 2025. General agentic memory via deep research. arXiv preprint arXiv:2511.18423. — [arXiv:2511.18423](https://arxiv.org/abs/2511.18423) find
  35. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  36. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William Cohen, Ruslan Salakhutdinov, and Christopher D Manning. 2018. Hotpotqa: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 conference on empirical methods in natural language processing, pages 2369–…
  37. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023. React: Synergizing reasoning and acting in language models. Preprint, arXiv:2210.03629.
  38. Ori Yoran, Tomer Wolfson, Ori Ram, and Jonathan Berant. 2023. Making retrieval-augmented language models robust to irrelevant context. arXiv preprint arXiv:2310.01558.
  39. Hongli Yu, Tinghong Chen, Jiangtao Feng, Jiangjie Chen, Weinan Dai, Qiying Yu, Ya-Qin Zhang, Wei-Ying Ma, Jingjing Liu, Mingxuan Wang, and 1 others. 2025. Memagent: Reshaping long-context llm with multi-conv rl-based memory agent. arXiv preprint arXiv:2507.02259. — [arXiv:2507.02259](https://arxiv.org/abs/2507.02259) find
  40. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. In International Conference on Learning Representations (ICLR). find
  41. Zeyu Zhang, Quanyu Dai, Xiaohe Bo, Chen Ma, Rui Li, Xu Chen, Jieming Zhu, Zhenhua Dong, and Ji-Rong Wen. 2025. A survey on the memory mechanism of large language model-based agents. ACM Transactions on Information Systems, 43(6):1–47. find
  42. Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, and Yanlin Wang. 2023. Memorybank: Enhancing large language models with long-term memory. Preprint, arXiv:2305.10250. find
  43. Shuyan Zhou, Frank F Xu, Hao Zhu, Xuhui Zhou, Robert Lo, Abishek Sridhar, Xianyi Cheng, Tianyue Ou, Yonatan Bisk, Daniel Fried, and 1 others. 2023. Webarena: A realistic web environment for building autonomous agents. arXiv preprint arXiv:2307.13854.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/roh-2026-clag-adaptive-memory-organization-agent/