---
title: "Mem²Evolve — paper by Cheng et al. (2026)"
description: "Others build new tools or agents, but they do so without lessons from earlier runs, so successes are… 56 references, cited by 5 papers here."
authors: ["Zihao Cheng", "Zeming Liu", "Yingyu Shan", "Xinyi Wang", "Xiangrong Zhu", "Yunpu Ma", "Hongru Wang", "Yuhang Guo", "Wei Lin", "Yunhong Wang"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.952/"
doi: "10.18653/v1/2026.acl-long.952"
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/cheng-2026-mem-2-evolve-self-evolving/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/cheng-2026-mem-2-evolve-self-evolving/)

[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/3/) / Mem²Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillation

# Mem²Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillation

**Authors:** Zihao Cheng, [Zeming Liu](https://md.research.elamur.ai/free/en/authors/zeming-liu/), [Yingyu Shan](https://md.research.elamur.ai/free/en/authors/yingyu-shan/), [Xinyi Wang](https://md.research.elamur.ai/free/en/authors/xinyi-wang/), Xiangrong Zhu, [Yunpu Ma](https://md.research.elamur.ai/free/en/authors/yunpu-ma/), [Hongru Wang](https://md.research.elamur.ai/free/en/authors/hongru-wang/), [Yuhang Guo](https://md.research.elamur.ai/free/en/authors/yuhang-guo/), [Wei Lin](https://md.research.elamur.ai/free/en/authors/wei-lin/) and Yunhong Wang · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/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.acl-long.952/) · [PDF](https://aclanthology.org/2026.acl-long.952.pdf) · [DOI 10.18653/v1/2026.acl-long.952](https://doi.org/10.18653/v1/2026.acl-long.952) · [OpenAlex](https://openalex.org/W7166844691)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [5](https://md.research.elamur.ai/free/en/papers/cheng-2026-mem-2-evolve-self-evolving/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
LLM-based self-evolving agents usually improve along one of two separate tracks. Some learn from accumulated experience but stay limited to a fixed, hand-defined set of tools and expert agents. Others build new tools or agents, but they do so without lessons from earlier runs, so successes are hard to repeat and errors recur. The authors argue these processes depend on each other and should evolve together. Their framework, Mem$^2$Evolve, takes inspiration from Piaget's equilibrium theory. It pairs an Asset Memory, which holds an Agent Bank of expert agents and a Tool Bank of executable tools, with an Experience Memory of lessons drawn from both successful and failed trajectories.

Each task runs in two phases:
- **Forward inference:** the agent reuses stored assets first and creates new ones, guided by experience, only when needed.
- **Backward evolution:** the agent keeps the useful new assets and records lessons that transfer to other tasks.

Across 6 task categories and 8 benchmarks, the approach beats plain LLMs, experience-only agents and creation-only agents. This makes it a more stable basis for lifelong-learning agents.

## Key points
- Average gains are +18.53% over standard LLMs, +11.80% over agents that evolve only through experience, and +6.46% over agents that evolve only through asset creation.
- The benchmarks cover six areas: general assistance (GAIA), multi-hop QA, math reasoning (AIME24/25), embodied tasks, planning and web interaction.
- Experience-guided tool creation raises average first-pass validity from 53.1% to 72.4%. On GAIA the relative gain is 56.0%, and on the AIME benchmarks fix iterations drop by roughly 68%.
- Memory built from GAIA transfers to 7 other benchmarks without negative transfer. It performs about as well as initializing with 25% same-task memory.
- A key limitation is that generated code must run in a sandbox. This restricts use in open-world settings that need local file system or unrestricted network access.

*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 Edinburgh](https://md.research.elamur.ai/free/en/universities/university-of-edinburgh/) — University, Edinburgh, United Kingdom · 359 works · 77,352 citations
- [Ludwig-Maximilians-Universität München](https://md.research.elamur.ai/free/en/universities/ludwig-maximilians-universitat-munchen/) — University, Munich, Germany · 107 works · 56,478 citations
- [Beihang University](https://md.research.elamur.ai/free/en/universities/beihang-university/) — University, Beijing, China · 65 works · 8,452 citations
- [Beijing Institute of Technology](https://md.research.elamur.ai/free/en/universities/beijing-institute-of-technology/) — University, Beijing, China · 34 works · 3,652 citations
- [Munich Center for Machine Learning](https://md.research.elamur.ai/free/en/universities/munich-center-for-machine-learning/) — University, Munich, Germany · 57 works · 1,094 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

## Cite

Cite as: Zihao Cheng, Zeming Liu, Yingyu Shan, Xinyi Wang, Xiangrong Zhu, Yunpu Ma, Hongru Wang, Yuhang Guo, Wei Lin and Yunhong Wang. 2026. Mem²Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillation. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{cheng2026self,
  title = {{Mem²Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillation}},
  author = {Zihao Cheng, Zeming Liu, Yingyu Shan, Xinyi Wang, Xiangrong Zhu, Yunpu Ma, Hongru Wang, Yuhang Guo, Wei Lin and Yunhong Wang},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.952},
  url = {https://doi.org/10.18653/v1/2026.acl-long.952}
}
```

- [← previous MedAraBench: Large-Scale Arabic Medical Question Answering Dataset and Benchmark](https://md.research.elamur.ai/free/en/papers/abu-daoud-2026-medarabench-large-scale-arabic-medical/)
- [next → MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction](https://md.research.elamur.ai/free/en/papers/cui-2026-minicpm-o-4-5/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/)

## References

56 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Sikai Bai, Haoxi Li, Jie Zhang, Yongjiang Liu, and Song Guo. 2026. Ttvs: Boosting self-exploring reinforcement learning via test-time variational synthesis. Preprint, arXiv:2604.08468. [find](https://scholar.google.com/scholar?q=Sikai+Bai%2C+Haoxi+Li%2C+Jie+Zhang%2C+Yongjiang+Liu%2C+and+Song+Guo%2E+2026%2E+Ttvs%3A+Boosting+self%2Dexploring+reinforcement+learning+via+test%2Dtime+variational+synthesis%2E+Preprint%2C+arXiv%3A2604%2E08468%2E)
2. Bytedance-Seed-Foundation-Code-Team, :, Yao Cheng, Jianfeng Chen, Jie Chen, Li Chen, Liyu Chen, Wentao Chen, Zhengyu Chen, Shijie Geng, Aoyan Li, Bo Li, Bowen Li, Linyi Li, Boyi Liu, Jiaheng Liu, Kaibo Liu, Qi Liu, Shukai Liu, and 37 others. 2025. Fullstack bench: Evaluating llms as full stack coder… — [arXiv:2412.00535](https://arxiv.org/abs/2412.00535) [find](https://scholar.google.com/scholar?q=Bytedance%2DSeed%2DFoundation%2DCode%2DTeam%2C+%3A%2C+Yao+Cheng%2C+Jianfeng+Chen%2C+Jie+Chen%2C+Li+Chen%2C+Liyu+Chen%2C+Wentao+Chen%2C+Zhengyu+Chen%2C+Shijie+Geng%2C+Aoyan+Li%2C+Bo+Li%2C+Bowen+Li%2C+Linyi+Li%2C+Boyi+Liu%2C+Jiaheng+Liu%2C+Kaib)
3. Tianle Cai, Xuezhi Wang, Tengyu Ma, Xinyun Chen, and Denny Zhou. 2024. Large language models as tool makers. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Tianle+Cai%2C+Xuezhi+Wang%2C+Tengyu+Ma%2C+Xinyun+Chen%2C+and+Denny+Zhou%2E+2024%2E+Large+language+models+as+tool+makers%2E+In+The+Twelfth+International+Conference+on+Learning+Representations%2E)
4. Mert Cemri, Melissa Z. Pan, Shuyi Yang, Lakshya A. Agrawal, Bhavya Chopra, Rishabh Tiwari, Kurt Keutzer, Aditya Parameswaran, Dan Klein, Kannan Ramchandran, Matei Zaharia, Joseph E. Gonzalez, and Ion Stoica. 2025. Why do multi-agent llm systems fail? Preprint, arXiv:2503.13657. [find](https://scholar.google.com/scholar?q=Mert+Cemri%2C+Melissa+Z%2E+Pan%2C+Shuyi+Yang%2C+Lakshya+A%2E+Agrawal%2C+Bhavya+Chopra%2C+Rishabh+Tiwari%2C+Kurt+Keutzer%2C+Aditya+Parameswaran%2C+Dan+Klein%2C+Kannan+Ramchandran%2C+Matei+Zaharia%2C+Joseph+E%2E+Gonzalez%2C+and+Ion+)
5. Guangyao Chen, Siwei Dong, Yu Shu, Ge Zhang, Jaward Sesay, Börje Karlsson, Jie Fu, and Yemin Shi. 2024. Autoagents: a framework for automatic agent generation. In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, pages 22–30. [find](https://scholar.google.com/scholar?q=Guangyao+Chen%2C+Siwei+Dong%2C+Yu+Shu%2C+Ge+Zhang%2C+Jaward+Sesay%2C+B%C3%B6rje+Karlsson%2C+Jie+Fu%2C+and+Yemin+Shi%2E+2024%2E+Autoagents%3A+a+framework+for+automatic+agent+generation%2E+In+Proceedings+of+the+Thirty%2DThird+Inter)
6. Weize Chen, Yusheng Su, Jingwei Zuo, Cheng Yang, Chenfei Yuan, Chi-Min Chan, Heyang Yu, Yaxi Lu, Yi-Hsin Hung, Chen Qian, and 1 others. 2023. [Agentverse: Facilitating multi-agent collaboration and exploring emergent behaviors](https://md.research.elamur.ai/free/en/papers/chen-2023-agentverse-facilitating-multi-agent/). In The Twelfth International Conference on Learning Representations.
7. Zihao Cheng, Yuheng Lu, Huaiqian Ye, Zeming Liu, Minqi Wang, Jingjing Liu, Zihan Li, Wei Fan, Yuan-fang Guo, Ruiji Fu, Shifeng She, Gang Wang, and Yunhong Wang. 2025a. Tcm-eval: An expert-level dynamic and extensible benchmark for traditional chinese medicine. Preprint, arXiv:2511.07148. — [arXiv:2511.07148](https://arxiv.org/abs/2511.07148) [find](https://scholar.google.com/scholar?q=Zihao+Cheng%2C+Yuheng+Lu%2C+Huaiqian+Ye%2C+Zeming+Liu%2C+Minqi+Wang%2C+Jingjing+Liu%2C+Zihan+Li%2C+Wei+Fan%2C+Yuan%2Dfang+Guo%2C+Ruiji+Fu%2C+Shifeng+She%2C+Gang+Wang%2C+and+Yunhong+Wang%2E+2025a%2E+Tcm%2Deval%3A+An+expert%2Dlevel+dynami)
8. Zihao Cheng, Hongru Wang, Zeming Liu, Yuhang Guo, Yuanfang Guo, Yunhong Wang, and Haifeng Wang. 2025b. ToolSpectrum: Towards personalized tool utilization for large language models. In Findings of the Association for Computational Linguistics: ACL 2025, pages 20679–20699, Vienna, Austria. Associatio… [find](https://scholar.google.com/scholar?q=Zihao+Cheng%2C+Hongru+Wang%2C+Zeming+Liu%2C+Yuhang+Guo%2C+Yuanfang+Guo%2C+Yunhong+Wang%2C+and+Haifeng+Wang%2E+2025b%2E+ToolSpectrum%3A+Towards+personalized+tool+utilization+for+large+language+models%2E+In+Findings+of+the)
9. Bin Deng, Yizhe Feng, Zeming Liu, Qing Wei, Xiangrong Zhu, Shuai Chen, Yuanfang Guo, and Yunhong Wang. 2025. RETAIL: Towards real-world travel planning for large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 14881–14913, Suzhou, Chi… [find](https://scholar.google.com/scholar?q=Bin+Deng%2C+Yizhe+Feng%2C+Zeming+Liu%2C+Qing+Wei%2C+Xiangrong+Zhu%2C+Shuai+Chen%2C+Yuanfang+Guo%2C+and+Yunhong+Wang%2E+2025%2E+RETAIL%3A+Towards+real%2Dworld+travel+planning+for+large+language+models%2E+In+Proceedings+of+the)
10. Jinyuan Fang, Yanwen Peng, Xi Zhang, Yingxu Wang, Xinhao Yi, Guibin Zhang, Yi Xu, Bin Wu, Siwei Liu, Zihao Li, Zhaochun Ren, Nikos Aletras, Xi Wang, Han Zhou, and Zaiqiao Meng. 2025. A comprehensive survey of self-evolving ai agents: A new paradigm bridging foundation models and lifelong agentic sys… — [arXiv:2508.07407](https://arxiv.org/abs/2508.07407) [find](https://scholar.google.com/scholar?q=Jinyuan+Fang%2C+Yanwen+Peng%2C+Xi+Zhang%2C+Yingxu+Wang%2C+Xinhao+Yi%2C+Guibin+Zhang%2C+Yi+Xu%2C+Bin+Wu%2C+Siwei+Liu%2C+Zihao+Li%2C+Zhaochun+Ren%2C+Nikos+Aletras%2C+Xi+Wang%2C+Han+Zhou%2C+and+Zaiqiao+Meng%2E+2025%2E+A+comprehensive+s)
11. Jiazhan Feng, Shijue Huang, Xingwei Qu, Ge Zhang, Yujia Qin, Baoquan Zhong, Chengquan Jiang, Jinxin Chi, and Wanjun Zhong. 2025. Retool: Reinforcement learning for strategic tool use in llms. Preprint, arXiv:2504.11536. [find](https://scholar.google.com/scholar?q=Jiazhan+Feng%2C+Shijue+Huang%2C+Xingwei+Qu%2C+Ge+Zhang%2C+Yujia+Qin%2C+Baoquan+Zhong%2C+Chengquan+Jiang%2C+Jinxin+Chi%2C+and+Wanjun+Zhong%2E+2025%2E+Retool%3A+Reinforcement+learning+for+strategic+tool+use+in+llms%2E+Preprint)
12. Communicative agents for" mind" exploration of large language model society. Advances in Neural Information Processing Systems, 36:51991–52008. Huan-ang Gao, Jiayi Geng, Wenyue Hua, Mengkang Hu, Xinzhe Juan, Hongzhang Liu, Shilong Liu, Jiahao Qiu, Xuan Qi, Yiran Wu, and 1 others. 2025. A survey of s… [find](https://scholar.google.com/scholar?q=Communicative+agents+for%22+mind%22+exploration+of+large+language+model+society%2E+Advances+in+Neural+Information+Processing+Systems%2C+36%3A51991%E2%80%9352008%2E+Huan%2Dang+Gao%2C+Jiayi+Geng%2C+Wenyue+Hua%2C+Mengkang+Hu%2C+Xinzh)
13. Xiaoxi Li, Wenxiang Jiao, Jiarui Jin, Guanting Dong, Jiajie Jin, Yinuo Wang, Hao Wang, Yutao Zhu, Ji-Rong Wen, Yuan Lu, and Zhicheng Dou. 2025b. Deepagent: A general reasoning agent with scalable toolsets. Preprint, arXiv:2510.21618. — [arXiv:2510.21618](https://arxiv.org/abs/2510.21618) [find](https://scholar.google.com/scholar?q=Xiaoxi+Li%2C+Wenxiang+Jiao%2C+Jiarui+Jin%2C+Guanting+Dong%2C+Jiajie+Jin%2C+Yinuo+Wang%2C+Hao+Wang%2C+Yutao+Zhu%2C+Ji%2DRong+Wen%2C+Yuan+Lu%2C+and+Zhicheng+Dou%2E+2025b%2E+Deepagent%3A+A+general+reasoning+agent+with+scalable+tool)
14. GitHub. 2025. Spec kit: Toolkit to help you get started with spec-driven development. [https://github.com/github/spec-kit](https://github.com/github/spec-kit). Accessed: 2025-12-19. [↗](https://github.com/github/spec-kit)
15. Jingjing Liu, Zeming Liu, Zihao Cheng, Mengliang He, Xiaoming Shi, Yuhang Guo, Xiangrong Zhu, Yuanfang Guo, Yunhong Wang, and Haifeng Wang. 2025a. RepoDebug: Repository-level multi-task and multi-language debugging evaluation of large language models. In Findings of the Association for Computational… [find](https://scholar.google.com/scholar?q=Jingjing+Liu%2C+Zeming+Liu%2C+Zihao+Cheng%2C+Mengliang+He%2C+Xiaoming+Shi%2C+Yuhang+Guo%2C+Xiangrong+Zhu%2C+Yuanfang+Guo%2C+Yunhong+Wang%2C+and+Haifeng+Wang%2E+2025a%2E+RepoDebug%3A+Repository%2Dlevel+multi%2Dtask+and+multi%2Dlang)
16. Ruiyan Han, Zhen Fang, XinYu Sun, Yuchen Ma, Ziheng Wang, Yu Zeng, Zehui Chen, Lin Chen, Wenxuan Huang, Wei-Jie Xu, and 1 others. 2026. [Unicorn: Towards self-improving unified multimodal models through self-generated supervision](https://md.research.elamur.ai/free/en/papers/fang-2026-unicorn-self-improving-unified-multimodal/). arXiv preprint arXiv:2601.03193.
17. Zhezheng Hao, Hong Wang, Jian Luo, Jianqing Zhang, Yuyan Zhou, Qiang Lin, Can Wang, Hande Dong, and Jiawei Chen. 2026. [Recreate: Reasoning and creating domain agents driven by experience](https://md.research.elamur.ai/free/en/papers/hao-2026-recreate-reasoning-creating-domain-agents/). arXiv preprint arXiv:2601.11100.
18. Xanh Ho, Anh-Khoa Duong Nguyen, Saku Sugawara, and Akiko Aizawa. 2020. [Constructing a multi-hop QA dataset for comprehensive evaluation of reasoning steps](https://md.research.elamur.ai/free/en/papers/ho-2020-constructing-multi-hop-qa-dataset/). In Proceedings of the 28th International Conference on Computational Linguistics, pages 6609–6625, Barcelona, Spain (Online). International Comm…
19. Ruili Jiang, Kehai Chen, Xuefeng Bai, Zhixuan He, Juntao Li, Muyun Yang, Tiejun Zhao, Liqiang Nie, and Min Zhang. 2025. A survey on human preference learning for aligning large language models. ACM Computing Surveys, 58(6):1–39. [find](https://scholar.google.com/scholar?q=Ruili+Jiang%2C+Kehai+Chen%2C+Xuefeng+Bai%2C+Zhixuan+He%2C+Juntao+Li%2C+Muyun+Yang%2C+Tiejun+Zhao%2C+Liqiang+Nie%2C+and+Min+Zhang%2E+2025%2E+A+survey+on+human+preference+learning+for+aligning+large+language+models%2E+ACM+Co)
20. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. [Search-r1: Training llms to reason and leverage search engines with reinforcement learning](https://md.research.elamur.ai/free/en/papers/jin-2025-search-r1-training-llms-reason/). arXiv preprint arXiv:2503.09516.
21. Omar Khattab, Arnav Singhvi, Paridhi Maheshwari, Zhiyuan Zhang, Keshav Santhanam, Sri Vardhamanan, Saiful Haq, Ashutosh Sharma, Thomas T Joshi, Hanna Moazam, and 1 others. 2023. [Dspy: Compiling declarative language model calls into self-improving pipelines](https://md.research.elamur.ai/free/en/papers/khattab-2023-dspy-compiling-declarative-language-model/). arXiv preprint arXiv:2310.03714.
22. Dawei Li, Bohan Jiang, Liangjie Huang, Alimohammad Beigi, Chengshuai Zhao, Zhen Tan, Amrita Bhattacharjee, Yuxuan Jiang, Canyu Chen, Tianhao Wu, and 1 others. 2025a. From generation to judgment: Opportunities and challenges of llm-as-a-judge. In Proceedings of the 2025 Conference on Empirical Method… [find](https://scholar.google.com/scholar?q=Dawei+Li%2C+Bohan+Jiang%2C+Liangjie+Huang%2C+Alimohammad+Beigi%2C+Chengshuai+Zhao%2C+Zhen+Tan%2C+Amrita+Bhattacharjee%2C+Yuxuan+Jiang%2C+Canyu+Chen%2C+Tianhao+Wu%2C+and+1+others%2E+2025a%2E+From+generation+to+judgment%3A+Oppor)
23. Guohao Li, Hasan Hammoud, Hani Itani, Dmitrii Khizbullin, and Bernard Ghanem. 2023. [find](https://scholar.google.com/scholar?q=Guohao+Li%2C+Hasan+Hammoud%2C+Hani+Itani%2C+Dmitrii+Khizbullin%2C+and+Bernard+Ghanem%2E+2023%2E)
24. Camel: Zeming Liu, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che, and Ting Liu. 2020. [Towards conversational recommendation over multi-type dialogs](https://md.research.elamur.ai/free/en/papers/liu-2020-conversational-recommendation-over-multi-type/). In Proceedings of the 58th annual meeting of the association for computational linguistics, pages 1036– 1049.
25. Zhipeng Liu, Xuefeng Bai, Kehai Chen, Xinyang Chen, Xiucheng Li, Yang Xiang, Jin Liu, Hong-Dong Li, Yaowei Wang, Liqiang Nie, and 1 others. 2025b. A survey on the feedback mechanism of llm-based ai agents. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence,… [find](https://scholar.google.com/scholar?q=Zhipeng+Liu%2C+Xuefeng+Bai%2C+Kehai+Chen%2C+Xinyang+Chen%2C+Xiucheng+Li%2C+Yang+Xiang%2C+Jin+Liu%2C+Hong%2DDong+Li%2C+Yaowei+Wang%2C+Liqiang+Nie%2C+and+1+others%2E+2025b%2E+A+survey+on+the+feedback+mechanism+of+llm%2Dbased+ai+ag)
26. Zijun Liu, Yanzhe Zhang, Peng Li, Yang Liu, and Diyi Yang. 2023. Dynamic llm-agent network: An llm-agent collaboration framework with agent team optimization. arXiv preprint arXiv:2310.02170. [find](https://scholar.google.com/scholar?q=Zijun+Liu%2C+Yanzhe+Zhang%2C+Peng+Li%2C+Yang+Liu%2C+and+Diyi+Yang%2E+2023%2E+Dynamic+llm%2Dagent+network%3A+An+llm%2Dagent+collaboration+framework+with+agent+team+optimization%2E+arXiv+preprint+arXiv%3A2310%2E02170%2E)
27. Yunbo Long, Yuhan Liu, and Liming Xu. 2026. [Emomas: Emotion-aware multi-agent system for high-stakes edge-deployable negotiation with bayesian orchestration](https://md.research.elamur.ai/free/en/papers/long-2026-emomas-emotion-aware-multi-agent/). Preprint, arXiv:2604.07003.
28. Kexin Ma, Bojun Li, Yuhua Tang, Liting Sun, and Ruochun Jin. 2026. Cast: Character-and-scene episodic memory for agents. Preprint, arXiv:2602.06051. — [arXiv:2602.06051](https://arxiv.org/abs/2602.06051) [find](https://scholar.google.com/scholar?q=Kexin+Ma%2C+Bojun+Li%2C+Yuhua+Tang%2C+Liting+Sun%2C+and+Ruochun+Jin%2E+2026%2E+Cast%3A+Character%2Dand%2Dscene+episodic+memory+for+agents%2E+Preprint%2C+arXiv%3A2602%2E06051%2E+%E2%80%94+%5BarXiv%3A2602%2E06051%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2602%2E060)
29. Xiaowen Ma, Chenyang Lin, Yao Zhang, Volker Tresp, and Yunpu Ma. 2025. Agentic neural networks: [Self-evolving multi-agent systems via textual backpropagation](https://md.research.elamur.ai/free/en/papers/ma-2026-self-evolving-multi-agent-systems/). arXiv preprint arXiv:2506.09046.
30. Grégoire Mialon, Clémentine Fourrier, Thomas Wolf, Yann LeCun, and Thomas Scialom. 2024. [GAIA: a benchmark for general AI assistants](https://md.research.elamur.ai/free/en/papers/mialon-2023-gaia-benchmark-general-ai-assistants/). In The Twelfth International Conference on Learning Representations.
31. OpenAI. 2025. Introducing deep research. [https://openai.com/index/](https://openai.com/index/) introducing-deep-research/. Accessed: 2026-01-02. [↗](https://openai.com/index/)
32. Siru Ouyang, Jun Yan, I Hsu, Yanfei Chen, Ke Jiang, Zifeng Wang, Rujun Han, Long T Le, Samira Daruki, Xiangru Tang, and 1 others. 2025. Reasoningbank: Scaling agent self-evolving with reasoning memory. arXiv preprint arXiv:2509.25140. — [arXiv:2509.25140](https://arxiv.org/abs/2509.25140) [find](https://scholar.google.com/scholar?q=Siru+Ouyang%2C+Jun+Yan%2C+I+Hsu%2C+Yanfei+Chen%2C+Ke+Jiang%2C+Zifeng+Wang%2C+Rujun+Han%2C+Long+T+Le%2C+Samira+Daruki%2C+Xiangru+Tang%2C+and+1+others%2E+2025%2E+Reasoningbank%3A+Scaling+agent+self%2Devolving+with+reasoning+memory)
33. Jean Piaget. 1972. Development and learning. Reading in child behavior and development, pages 38–46. [find](https://scholar.google.com/scholar?q=Jean+Piaget%2E+1972%2E+Development+and+learning%2E+Reading+in+child+behavior+and+development%2C+pages+38%E2%80%9346%2E)
34. Jiahao Qiu, Xuan Qi, Tongcheng Zhang, Xinzhe Juan, Jiacheng Guo, Yifu Lu, Yimin Wang, Zixin Yao, Qihan Ren, Xun Jiang, and 1 others. 2025. Alita: Generalist agent enabling scalable agentic reasoning with minimal predefinition and maximal self-evolution. arXiv preprint arXiv:2505.20286. — [arXiv:2505.20286](https://arxiv.org/abs/2505.20286) [find](https://scholar.google.com/scholar?q=Jiahao+Qiu%2C+Xuan+Qi%2C+Tongcheng+Zhang%2C+Xinzhe+Juan%2C+Jiacheng+Guo%2C+Yifu+Lu%2C+Yimin+Wang%2C+Zixin+Yao%2C+Qihan+Ren%2C+Xun+Jiang%2C+and+1+others%2E+2025%2E+Alita%3A+Generalist+agent+enabling+scalable+agentic+reasoning+w)
35. Deng Qiyuan, Kehai Chen, Min Zhang, and Zhongwen Xu. 2026. HiPO: Self-hint policy optimization for RLVR. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Deng+Qiyuan%2C+Kehai+Chen%2C+Min+Zhang%2C+and+Zhongwen+Xu%2E+2026%2E+HiPO%3A+Self%2Dhint+policy+optimization+for+RLVR%2E+In+The+Fourteenth+International+Conference+on+Learning+Representations%2E)
36. Yu Shang, Yu Li, Keyu Zhao, Likai Ma, Jiahe Liu, Fengli Xu, and Yong Li. 2025. Agentsquare: Automatic LLM agent search in modular design space. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Yu+Shang%2C+Yu+Li%2C+Keyu+Zhao%2C+Likai+Ma%2C+Jiahe+Liu%2C+Fengli+Xu%2C+and+Yong+Li%2E+2025%2E+Agentsquare%3A+Automatic+LLM+agent+search+in+modular+design+space%2E+In+The+Thirteenth+International+Conference+on+Learning+R)
37. Mohit Shridhar, Xingdi Yuan, Marc-Alexandre Côté, Yonatan Bisk, Adam Trischler, and Matthew Hausknecht. 2020. [Alfworld: Aligning text and embodied environments for interactive learning](https://md.research.elamur.ai/free/en/papers/shridhar-2020-alfworld-aligning-text-embodied-environments/). arXiv preprint arXiv:2010.03768.
38. Linxin Song, Jiale Liu, Jieyu Zhang, Shaokun Zhang, Ao Luo, Shijian Wang, Qingyun Wu, and Chi Wang. 2024. Adaptive in-conversation team building for language model agents. arXiv preprint arXiv:2405.19425. — [arXiv:2405.19425](https://arxiv.org/abs/2405.19425) [find](https://scholar.google.com/scholar?q=Linxin+Song%2C+Jiale+Liu%2C+Jieyu+Zhang%2C+Shaokun+Zhang%2C+Ao+Luo%2C+Shijian+Wang%2C+Qingyun+Wu%2C+and+Chi+Wang%2E+2024%2E+Adaptive+in%2Dconversation+team+building+for+language+model+agents%2E+arXiv+preprint+arXiv%3A2405%2E19)
39. UncleCode. 2024. Crawl4ai: Open-source llm friendly web crawler & scraper. [https://github.com/](https://github.com/) unclecode/crawl4ai. [↗](https://github.com/)
40. Guancheng Wan, Mo Zhou, Ziyi Wang, Xiaoran Shang, Eric Hanchen Jiang, Guibin Zhang, Jinhe Bi, Yunpu Ma, Zaixi Zhang, Ke Liang, and 1 others. 2026. Dawn: Distributed llm multi-agent workflow synthesis. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 26099–26106. [find](https://scholar.google.com/scholar?q=Guancheng+Wan%2C+Mo+Zhou%2C+Ziyi+Wang%2C+Xiaoran+Shang%2C+Eric+Hanchen+Jiang%2C+Guibin+Zhang%2C+Jinhe+Bi%2C+Yunpu+Ma%2C+Zaixi+Zhang%2C+Ke+Liang%2C+and+1+others%2E+2026%2E+Dawn%3A+Distributed+llm+multi%2Dagent+workflow+synthesis%2E)
41. Hongru Wang, Cheng Qian, Manling Li, Jiahao Qiu, Boyang Xue, Mengdi Wang, Heng Ji, and Kam-Fai Wong. 2025a. Toward a theory of agents as tool-use decision-makers. arXiv preprint arXiv:2506.00886. — [arXiv:2506.00886](https://arxiv.org/abs/2506.00886) [find](https://scholar.google.com/scholar?q=Hongru+Wang%2C+Cheng+Qian%2C+Manling+Li%2C+Jiahao+Qiu%2C+Boyang+Xue%2C+Mengdi+Wang%2C+Heng+Ji%2C+and+Kam%2DFai+Wong%2E+2025a%2E+Toward+a+theory+of+agents+as+tool%2Duse+decision%2Dmakers%2E+arXiv+preprint+arXiv%3A2506%2E00886%2E+%E2%80%94+%5Ba)
42. Hongru Wang, Cheng Qian, Wanjun Zhong, Xiusi Chen, Jiahao Qiu, Shijue Huang, Bowen Jin, Mengdi Wang, Kam-Fai Wong, and Heng Ji. 2025b. Acting less is reasoning more! teaching model to act efficiently. arXiv preprint arXiv:2504.14870. — [arXiv:2504.14870](https://arxiv.org/abs/2504.14870) [find](https://scholar.google.com/scholar?q=Hongru+Wang%2C+Cheng+Qian%2C+Wanjun+Zhong%2C+Xiusi+Chen%2C+Jiahao+Qiu%2C+Shijue+Huang%2C+Bowen+Jin%2C+Mengdi+Wang%2C+Kam%2DFai+Wong%2C+and+Heng+Ji%2E+2025b%2E+Acting+less+is+reasoning+more%21+teaching+model+to+act+efficiently%2E)
43. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). Advances in neural information processing systems, 35:24824– 24837.
44. Georg Wölflein, Dyke Ferber, Daniel Truhn, Ognjen Arandjelovic, and Jakob Nikolas Kather. 2025. [LLM agents making agent tools](https://md.research.elamur.ai/free/en/papers/wolflein-2025-llm-agents-making-agent-tools/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 26092–26130, Vienna, Austria. Association for Compu…
45. Jian Xie, Kai Zhang, Jiangjie Chen, Tinghui Zhu, Renze Lou, Yuandong Tian, Yanghua Xiao, and Yu Su. 2024. Travelplanner: a benchmark for real-world planning with language agents. In Proceedings of the 41st International Conference on Machine Learning, pages 54590–54613. [find](https://scholar.google.com/scholar?q=Jian+Xie%2C+Kai+Zhang%2C+Jiangjie+Chen%2C+Tinghui+Zhu%2C+Renze+Lou%2C+Yuandong+Tian%2C+Yanghua+Xiao%2C+and+Yu+Su%2E+2024%2E+Travelplanner%3A+a+benchmark+for+real%2Dworld+planning+with+language+agents%2E+In+Proceedings+of+the)
46. John Yang, Carlos E Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press. 2024. [Swe-agent: Agent-computer interfaces enable automated software engineering](https://md.research.elamur.ai/free/en/papers/yang-2024-swe-agent-agent-computer-interfaces/). Advances in Neural Information Processing Systems, 37:50528– 50652.
47. 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](https://md.research.elamur.ai/free/en/papers/yang-2018-hotpotqa-dataset-diverse-explainable-multi/). In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2369…
48. Shunyu Yao, Howard Chen, John Yang, and Karthik Narasimhan. 2022a. [Webshop: Towards scalable real-world web interaction with grounded language agents](https://md.research.elamur.ai/free/en/papers/yao-2022-webshop-towards-scalable-real-world/). Advances in Neural Information Processing Systems, 35:20744–20757.
49. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. 2022b. [React: Synergizing reasoning and acting in language models](https://md.research.elamur.ai/free/en/papers/yao-2022-react-synergizing-reasoning-acting-language/). In The eleventh international conference on learning representations. yt-dlp. 2025. yt-dlp: A feature-rich command-line audio/video downloa…
50. Siyu Yuan, Kaitao Song, Jiangjie Chen, Xu Tan, Dongsheng Li, and Deqing Yang. 2025. Evoagent: Towards automatic multi-agent generation via evolutionary algorithms. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Lan… [find](https://scholar.google.com/scholar?q=Siyu+Yuan%2C+Kaitao+Song%2C+Jiangjie+Chen%2C+Xu+Tan%2C+Dongsheng+Li%2C+and+Deqing+Yang%2E+2025%2E+Evoagent%3A+Towards+automatic+multi%2Dagent+generation+via+evolutionary+algorithms%2E+In+Proceedings+of+the+2025+Conferenc)
51. Liang Yue, Yihong Tang, Kehai Chen, Jie Liu, and Min Zhang. 2025. Master: Enhancing large language model via multi-agent simulated teaching. arXiv preprint arXiv:2506.02689. — [arXiv:2506.02689](https://arxiv.org/abs/2506.02689) [find](https://scholar.google.com/scholar?q=Liang+Yue%2C+Yihong+Tang%2C+Kehai+Chen%2C+Jie+Liu%2C+and+Min+Zhang%2E+2025%2E+Master%3A+Enhancing+large+language+model+via+multi%2Dagent+simulated+teaching%2E+arXiv+preprint+arXiv%3A2506%2E02689%2E+%E2%80%94+%5BarXiv%3A2506%2E02689%5D%28https)
52. Mert Yuksekgonul, Federico Bianchi, Joseph Boen, Sheng Liu, Pan Lu, Zhi Huang, Carlos Guestrin, and James Zou. 2025. Optimizing generative ai by backpropagating language model feedback. Nature, 639(8055):609–616. [find](https://scholar.google.com/scholar?q=Mert+Yuksekgonul%2C+Federico+Bianchi%2C+Joseph+Boen%2C+Sheng+Liu%2C+Pan+Lu%2C+Zhi+Huang%2C+Carlos+Guestrin%2C+and+James+Zou%2E+2025%2E+Optimizing+generative+ai+by+backpropagating+language+model+feedback%2E+Nature%2C+639%2880)
53. Jiayi Zhang, Jinyu Xiang, Zhaoyang Yu, Fengwei Teng, Xiong-Hui Chen, Jiaqi Chen, Mingchen Zhuge, Xin Cheng, Sirui Hong, Jinlin Wang, and 1 others. 2025a. Aflow: Automating agentic workflow generation. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Jiayi+Zhang%2C+Jinyu+Xiang%2C+Zhaoyang+Yu%2C+Fengwei+Teng%2C+Xiong%2DHui+Chen%2C+Jiaqi+Chen%2C+Mingchen+Zhuge%2C+Xin+Cheng%2C+Sirui+Hong%2C+Jinlin+Wang%2C+and+1+others%2E+2025a%2E+Aflow%3A+Automating+agentic+workflow+generation%2E)
54. Wenyuan Zhang, Xinghua Zhang, Haiyang Yu, Shuaiyi Nie, Bingli Wu, Juwei Yue, Tingwen Liu, and Yongbin Li. 2026. [Expseek: Self-triggered experience seeking for web agents](https://md.research.elamur.ai/free/en/papers/zhang-2026-expseek-self-triggered-experience-seeking/). Preprint, arXiv:2601.08605.
55. Yao Zhang, Chenyang Lin, Shijie Tang, Haokun Chen, Shijie Zhou, Yunpu Ma, and Volker Tresp. 2025b. Swarmagentic: Towards fully automated agentic system generation via swarm intelligence. arXiv preprint arXiv:2506.15672. [find](https://scholar.google.com/scholar?q=Yao+Zhang%2C+Chenyang+Lin%2C+Shijie+Tang%2C+Haokun+Chen%2C+Shijie+Zhou%2C+Yunpu+Ma%2C+and+Volker+Tresp%2E+2025b%2E+Swarmagentic%3A+Towards+fully+automated+agentic+system+generation+via+swarm+intelligence%2E+arXiv+preprin)
56. Hongli Zhou, Hui Huang, Ziqing Zhao, Lvyuan Han, Huicheng Wang, Kehai Chen, Muyun Yang, Wei Bao, Jian Dong, Bing Xu, and 1 others. 2026. Lost in benchmarks? rethinking large language model benchmarking with item response theory. In Proceedings of the AAAI Conference on Artificial Intelligence, volum… [find](https://scholar.google.com/scholar?q=Hongli+Zhou%2C+Hui+Huang%2C+Ziqing+Zhao%2C+Lvyuan+Han%2C+Huicheng+Wang%2C+Kehai+Chen%2C+Muyun+Yang%2C+Wei+Bao%2C+Jian+Dong%2C+Bing+Xu%2C+and+1+others%2E+2026%2E+Lost+in+benchmarks%3F+rethinking+large+language+model+benchmarkin)

*Updated 1 November 2026*