---
title: "LLM-MemCluster — paper by Zhu et al. (2026)"
description: "It may also propose merging similar labels, after which earlier assignments are relabelled. 26 references."
authors: ["Yuanjie Zhu", "Liangwei Yang", "Ke Xu", "Weizhi Zhang", "Zihe Song", "Jindong Wang", "Philip S. Yu"]
year: 2026
venue: "Proceedings of the 4th Workshop on Towards Knowledgeable Foundation Models (KnowFM 2026)"
source: "https://aclanthology.org/2026.knowfm-1.7/"
doi: "10.18653/v1/2026.knowfm-1.7"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/zhu-2026-llm-memcluster-empowering-large-language/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhu-2026-llm-memcluster-empowering-large-language/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/39/) / LLM-MemCluster: Empowering Large Language Models with Dynamic Memory for Text Clustering

# LLM-MemCluster: Empowering Large Language Models with Dynamic Memory for Text Clustering

**Authors:** Yuanjie Zhu, Liangwei Yang, [Ke Xu](https://md.research.elamur.ai/free/en/authors/ke-xu/), [Weizhi Zhang](https://md.research.elamur.ai/free/en/authors/weizhi-zhang/), Zihe Song, [Jindong Wang](https://md.research.elamur.ai/free/en/authors/jindong-wang/) and [Philip S. Yu](https://md.research.elamur.ai/free/en/authors/philip-s-yu/) · **Year:** 2026 · **Venue:** [Proceedings of the 4th Workshop on Towards Knowledgeable Foundation Models (KnowFM 2026)](https://md.research.elamur.ai/free/en/venues/knowfm/) ([Workshop on Towards Knowledgeable Foundation Models 2026](https://md.research.elamur.ai/free/en/venues/knowfm/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.knowfm-1.7/) · [PDF](https://aclanthology.org/2026.knowfm-1.7.pdf) · [DOI 10.18653/v1/2026.knowfm-1.7](https://doi.org/10.18653/v1/2026.knowfm-1.7) · [OpenAlex](https://openalex.org/W7166813346)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
LLMs can group texts by meaning, but using them directly as clustering engines runs into two obstacles. Large corpora must be fed in pieces, and the model forgets earlier decisions between calls. The model also has no built-in way to decide how coarse or fine the clusters should be. Most prior work therefore keeps LLMs in a supporting role inside embedding-based pipelines.

LLM-MemCluster makes clustering fully LLM-driven, with no fine-tuning and no classical algorithm. It reads documents one at a time and keeps a prompt-resident memory of current cluster labels. For each text, the LLM either reuses a label or invents a new one. It may also propose merging similar labels, after which earlier assignments are relabelled. A dual-prompt switch controls granularity: once the label count reaches a user-set upper bound, a strict prompt discourages new labels; otherwise a relaxed prompt allows them.

On six public benchmarks, the authors report consistent gains over embedding-based and LLM-augmented baselines across several LLMs. The approach is interpretable and needs no training.

## Key points
- **Single pass:** the method clusters N documents in exactly N LLM calls. Each call returns a label plus an optional merge suggestion, e.g. folding {"ML", "DL"} into "AI".
- **Mode switching:** the user supplies a target range [K_min, K_max]. The strict mode activates when the current label count is at or above K_max; otherwise the relaxed mode is used.
- **Evaluation setup:**
  - six benchmark datasets taken from Zhang et al. (2023), with K from 18 to 102;
  - three standard metrics, where 1 is a perfect score;
  - robustness checks on the switching threshold and on dataset order;
  - tests with both proprietary and open-source LLMs.
- **Related systems:** the paper positions itself against ClusterLLM, LLMEdgeRefine, Cequel, T-CLC and ZeroDL. Cost is driven mainly by input tokens, while output tokens stay minimal.
- **Limitations:**
  - it depends on strong instruction following, so smaller models may struggle;
  - it performs no multi-pass global refinement;
  - users need rough prior knowledge of the expected cluster count.

*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

- [University of Illinois Chicago](https://md.research.elamur.ai/free/en/universities/university-of-illinois-chicago/) — University, Chicago, United States · 70 works · 20,363 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🤝 General Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/general-social-sciences/) · 254 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works

## Cite

Cite as: Yuanjie Zhu, Liangwei Yang, Ke Xu, Weizhi Zhang, Zihe Song, Jindong Wang and Philip S. Yu. 2026. LLM-MemCluster: Empowering Large Language Models with Dynamic Memory for Text Clustering. Proceedings of the 4th Workshop on Towards Knowledgeable Foundation Models (KnowFM 2026).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{zhu2026llm,
  title = {{LLM-MemCluster: Empowering Large Language Models with Dynamic Memory for Text Clustering}},
  author = {Yuanjie Zhu, Liangwei Yang, Ke Xu, Weizhi Zhang, Zihe Song, Jindong Wang and Philip S. Yu},
  booktitle = {Proceedings of the 4th Workshop on Towards Knowledgeable Foundation Models (KnowFM 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.knowfm-1.7},
  url = {https://doi.org/10.18653/v1/2026.knowfm-1.7}
}
```

- [← previous LLM-Generated Stories for Students with Significant Cognitive Disabilities: Promise, Gaps, and Evaluation Framework](https://md.research.elamur.ai/free/en/papers/maheshwary-2026-llm-generated-stories-students-significant/)
- [next → LLM-VA: Resolving the Jailbreak-Overrefusal Trade-off via Vector Alignment](https://md.research.elamur.ai/free/en/papers/zhang-2026-llm-va-resolving-jailbreak-overrefusal/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/39/)

## References

26 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Tianyu Gao, Xu Han, Hao Zhu, Zhiyuan Liu, Peng Li, Maosong Sun, and Jie Zhou. 2019. FewRel 2.0: Towards more challenging few-shot relation classification. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural L… [find](https://scholar.google.com/scholar?q=Tianyu+Gao%2C+Xu+Han%2C+Hao+Zhu%2C+Zhiyuan+Liu%2C+Peng+Li%2C+Maosong+Sun%2C+and+Jie+Zhou%2E+2019%2E+FewRel+2%2E0%3A+Towards+more+challenging+few%2Dshot+relation+classification%2E+In+Proceedings+of+the+2019+Conference+on+Empi)
2. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774.
3. Prafulla Bafna, Dhanya Pramod, and Anagha Vaidya. 2016. Document clustering: Tf-idf approach. In 2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT), pages 61–66. [find](https://scholar.google.com/scholar?q=Prafulla+Bafna%2C+Dhanya+Pramod%2C+and+Anagha+Vaidya%2E+2016%2E+Document+clustering%3A+Tf%2Didf+approach%2E+In+2016+International+Conference+on+Electrical%2C+Electronics%2C+and+Optimization+Techniques+%28ICEEOT%29%2C+pages+6)
4. Maarten Grootendorst. 2022. [Bertopic: Neural topic modeling with a class-based tf-idf procedure](https://md.research.elamur.ai/free/en/papers/grootendorst-2022-bertopic-neural-topic-modeling-class/). arXiv preprint arXiv:2203.05794.
5. Sugato Basu, Arindam Banerjee, and Raymond J Mooney. 2004. Active semi-supervision for pairwise constrained clustering. In Proceedings of the 2004 SIAM international conference on data mining, pages 333–344. SIAM. [find](https://scholar.google.com/scholar?q=Sugato+Basu%2C+Arindam+Banerjee%2C+and+Raymond+J+Mooney%2E+2004%2E+Active+semi%2Dsupervision+for+pairwise+constrained+clustering%2E+In+Proceedings+of+the+2004+SIAM+international+conference+on+data+mining%2C+pages+3)
6. Amir Hadifar, Lucas Sterckx, Thomas Demeester, and Chris Develder. 2019. A self-training approach for short text clustering. In Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP-2019), pages 194–199, Florence, Italy. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Amir+Hadifar%2C+Lucas+Sterckx%2C+Thomas+Demeester%2C+and+Chris+Develder%2E+2019%2E+A+self%2Dtraining+approach+for+short+text+clustering%2E+In+Proceedings+of+the+4th+Workshop+on+Representation+Learning+for+NLP+%28RepL)
7. Dingsheng Deng. 2020. Dbscan clustering algorithm based on density. In 2020 7th international forum on electrical engineering and automation (IFEEA), pages 949–953. IEEE. [find](https://scholar.google.com/scholar?q=Dingsheng+Deng%2E+2020%2E+Dbscan+clustering+algorithm+based+on+density%2E+In+2020+7th+international+forum+on+electrical+engineering+and+automation+%28IFEEA%29%2C+pages+949%E2%80%93953%2E+IEEE%2E)
8. Chen Huang and Guoxiu He. 2025. Text clustering as classification with llms. In Proceedings of the 2025 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region, SIGIR-AP 2025, page 374–384, New York, NY, USA. Association for Computing… [find](https://scholar.google.com/scholar?q=Chen+Huang+and+Guoxiu+He%2E+2025%2E+Text+clustering+as+classification+with+llms%2E+In+Proceedings+of+the+2025+Annual+International+ACM+SIGIR+Conference+on+Research+and+Development+in+Information+Retrieval+i)
9. Ning Ding, Guangwei Xu, Yulin Chen, Xiaobin Wang, Xu Han, Pengjun Xie, Haitao Zheng, and Zhiyuan Liu. 2021. [Few-NERD: A few-shot named entity recognition dataset](https://md.research.elamur.ai/free/en/papers/ding-2021-few-nerd-few-shot-named/). In Proceedings of the 59th Annual Meeting of the Association for Computational Xin Jin and Jiawei Han. 2017. K-means clustering. In Encyc…
10. Kristina P Sinaga and Miin-Shen Yang. 2020. Unsupervised k-means clustering algorithm. IEEE access, 8:80716–80727. [find](https://scholar.google.com/scholar?q=Kristina+P+Sinaga+and+Miin%2DShen+Yang%2E+2020%2E+Unsupervised+k%2Dmeans+clustering+algorithm%2E+IEEE+access%2C+8%3A80716%E2%80%9380727%2E)
11. Hwiyeol Jo, Hyunwoo Lee, Kang Min Yoo, and Taiwoo Park. 2025. ZeroDL: Zero-shot distribution learning for text clustering via large language models. In Findings of the Association for Computational Linguistics: ACL 2025, pages 19597–19607, Vienna, Austria. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Hwiyeol+Jo%2C+Hyunwoo+Lee%2C+Kang+Min+Yoo%2C+and+Taiwoo+Park%2E+2025%2E+ZeroDL%3A+Zero%2Dshot+distribution+learning+for+text+clustering+via+large+language+models%2E+In+Findings+of+the+Association+for+Computational+Li)
12. Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, and Tao Yu. 2023. [One embedder, any task: Instruction-finetuned text embeddings](https://md.research.elamur.ai/free/en/papers/su-2022-one-embedder-any-task-instruction/). In Findings of the Association for Computational Linguistics: ACL 2023, pages 1102–1121, Toronto…
13. Haoran Li, Abhinav Arora, Shuohui Chen, Anchit Gupta, Sonal Gupta, and Yashar Mehdad. 2021. [MTOP: A comprehensive multilingual task-oriented semantic parsing benchmark](https://md.research.elamur.ai/free/en/papers/li-2021-mtop-comprehensive-multilingual-task-oriented/). In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages…
14. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, and 1 others. 2023. [Gemini: a family of highly capable multimodal models](https://md.research.elamur.ai/free/en/papers/gemini-team-2023-gemini-family-highly-capable-multimodal/). arXiv preprint arXiv:2312.11805.
15. Hongtao Wang, Taiyan Zhang, Renchi Yang, and Jianliang Xu. 2025. Cequel: Cost-effective querying of large language models for text clustering. In Proceedings of the 34th ACM International Conference on Information and Knowledge Management, CIKM ’25, page 2998–3008, New York, NY, USA. Association for… [find](https://scholar.google.com/scholar?q=Hongtao+Wang%2C+Taiyan+Zhang%2C+Renchi+Yang%2C+and+Jianliang+Xu%2E+2025%2E+Cequel%3A+Cost%2Deffective+querying+of+large+language+models+for+text+clustering%2E+In+Proceedings+of+the+34th+ACM+International+Conference+o)
16. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+and+1+others%2E+2024%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A2412)
17. Guangda Liu, Chengwei Li, Jieru Zhao, Chenqi Zhang, and Minyi Guo. 2025. Clusterkv: Manipulating llm kv cache in semantic space for recallable compression. In 2025 62nd ACM/IEEE Design Automation Conference (DAC), pages 1–7. [find](https://scholar.google.com/scholar?q=Guangda+Liu%2C+Chengwei+Li%2C+Jieru+Zhao%2C+Chenqi+Zhang%2C+and+Minyi+Guo%2E+2025%2E+Clusterkv%3A+Manipulating+llm+kv+cache+in+semantic+space+for+recallable+compression%2E+In+2025+62nd+ACM%2FIEEE+Design+Automation+Conf)
18. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, and Furu Wei. 2024. [Improving text embeddings with large language models](https://md.research.elamur.ai/free/en/papers/wang-2024-improving-text-embeddings-large-language/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11897–11916, Bangkok, Thailand. Asso…
19. Stuart Lloyd. 1982. Least squares quantization in pcm. IEEE transactions on information theory, 28(2):129– 137. [find](https://scholar.google.com/scholar?q=Stuart+Lloyd%2E+1982%2E+Least+squares+quantization+in+pcm%2E+IEEE+transactions+on+information+theory%2C+28%282%29%3A129%E2%80%93+137%2E)
20. Wujiang Xu, Kai Mei, Hang Gao, Juntao Tan, Zujie Liang, and Yongfeng Zhang. 2025. [A-mem: Agentic memory for llm agents](https://md.research.elamur.ai/free/en/papers/xu-2025-mem-agentic-memory-llm-agents/). arXiv preprint arXiv:2502.12110.
21. Niklas Muennighoff, Nouamane Tazi, Loic Magne, and Nils Reimers. 2023. [MTEB: Massive text embedding benchmark](https://md.research.elamur.ai/free/en/papers/muennighoff-2022-mteb-massive-text-embedding-benchmark/). In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 2014–2037, Dubrovnik, Croatia. Association for Computational Linguistic…
22. Yuwei Zhang, Zihan Wang, and Jingbo Shang. 2023. ClusterLLM: Large language models as a guide for text clustering. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 13903–13920, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Yuwei+Zhang%2C+Zihan+Wang%2C+and+Jingbo+Shang%2E+2023%2E+ClusterLLM%3A+Large+language+models+as+a+guide+for+text+clustering%2E+In+Proceedings+of+the+2023+Conference+on+Empirical+Methods+in+Natural+Language+Proces)
23. Andrew Ng, Michael Jordan, and Yair Weiss. 2001. On spectral clustering: Analysis and an algorithm. Advances in neural information processing systems, 14. [find](https://scholar.google.com/scholar?q=Andrew+Ng%2C+Michael+Jordan%2C+and+Yair+Weiss%2E+2001%2E+On+spectral+clustering%3A+Analysis+and+an+algorithm%2E+Advances+in+neural+information+processing+systems%2C+14%2E)
24. Sheng Zhou, Hongjia Xu, Zhuonan Zheng, Jiawei Chen, Zhao Li, Jiajun Bu, Jia Wu, Xin Wang, Wenwu Zhu, and Martin Ester. 2024. A comprehensive survey on deep clustering: Taxonomy, challenges, and future directions. ACM Comput. Surv., 57(3). [find](https://scholar.google.com/scholar?q=Sheng+Zhou%2C+Hongjia+Xu%2C+Zhuonan+Zheng%2C+Jiawei+Chen%2C+Zhao+Li%2C+Jiajun+Bu%2C+Jia+Wu%2C+Xin+Wang%2C+Wenwu+Zhu%2C+and+Martin+Ester%2E+2024%2E+A+comprehensive+survey+on+deep+clustering%3A+Taxonomy%2C+challenges%2C+and+future)
25. Gabor Petnehazi and Bernadett Aradi. 2025. Hercules: Hierarchical embedding-based recursive clustering using llms for efficient summarization. arXiv preprint arXiv:2506.19992. — [arXiv:2506.19992](https://arxiv.org/abs/2506.19992) [find](https://scholar.google.com/scholar?q=Gabor+Petnehazi+and+Bernadett+Aradi%2E+2025%2E+Hercules%3A+Hierarchical+embedding%2Dbased+recursive+clustering+using+llms+for+efficient+summarization%2E+arXiv+preprint+arXiv%3A2506%2E19992%2E+%E2%80%94+%5BarXiv%3A2506%2E19992%5D%28htt)
26. Xingcheng Ran, Yue Xi, Yonggang Lu, Xiangwen Wang, and Zhenyu Lu. 2023. Comprehensive survey on hierarchical clustering algorithms and the recent developments. Artificial Intelligence Review, 56(8):8219–8264. [find](https://scholar.google.com/scholar?q=Xingcheng+Ran%2C+Yue+Xi%2C+Yonggang+Lu%2C+Xiangwen+Wang%2C+and+Zhenyu+Lu%2E+2023%2E+Comprehensive+survey+on+hierarchical+clustering+algorithms+and+the+recent+developments%2E+Artificial+Intelligence+Review%2C+56%288%29%3A82)

*Updated 2 November 2026*