---
title: "DarwinTOD — paper by Zhang et al. (2026) — summary and ideas"
description: "Pipeline designs pass errors from one module to the next, end-to-end LLM systems do not learn from use, and continual… 56 references."
authors: ["Shuyu Zhang", "Yujie Liu", "Xinru Wang", "Cheng Zhang", "Yanmin Zhu", "Bin Li"]
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.2050/"
doi: "10.18653/v1/2026.acl-long.2050"
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/zhang-2026-darwintod-llm-driven-lifelong-self/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhang-2026-darwintod-llm-driven-lifelong-self/)

[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/29/) / DarwinTOD: LLM-Driven Lifelong Self-evolution for Task-oriented Dialog Systems

# DarwinTOD: LLM-Driven Lifelong Self-evolution for Task-oriented Dialog Systems

**Authors:** [Shuyu Zhang](https://md.research.elamur.ai/free/en/authors/shuyu-zhang/), [Yujie Liu](https://md.research.elamur.ai/free/en/authors/yujie-liu/), [Xinru Wang](https://md.research.elamur.ai/free/en/authors/xinru-wang/), [Cheng Zhang](https://md.research.elamur.ai/free/en/authors/cheng-zhang/), [Yanmin Zhu](https://md.research.elamur.ai/free/en/authors/yanmin-zhu/) and [Bin Li](https://md.research.elamur.ai/free/en/authors/bin-li/) · **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.2050/) · [PDF](https://aclanthology.org/2026.acl-long.2050.pdf) · [DOI 10.18653/v1/2026.acl-long.2050](https://doi.org/10.18653/v1/2026.acl-long.2050) · [OpenAlex](https://openalex.org/W7166827212)

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

## Summary
Task-oriented dialog (TOD) systems usually stay frozen after deployment. Pipeline designs pass errors from one module to the next, end-to-end LLM systems do not learn from use, and continual learning still depends on periodic retraining with curated data. DarwinTOD tackles this by letting a TOD system improve itself across its whole lifetime. It starts from a zero-shot LLM and needs no task-specific fine-tuning or human labeling.

The core is an Evolvable Strategy Bank (ESB), a population of dialog strategies that compete and are chosen by fitness. Two loops alternate:
- **Online loop:** several agents (DST, DP, NLG, and a user simulator) carry out dialogs and critique one another.
- **Offline loop:** LLM-driven evolutionary operators mutate, consolidate and prune strategies using the collected feedback.

The authors frame dialog as a POMDP and the bank's development as a Markov chain. On MultiWOZ and SGD, they report beating prior state-of-the-art methods, with steady gains over generations. The bank also organizes itself into domain-specialized strategies. The work points toward dialog agents that keep adapting after release without manual retraining.

## Key points
- **Fitness score:** each strategy's score combines its net positive-minus-negative feedback per use with an age term based on a min-max normalized generation index. Feedback comes from task success and peer critiques. The weight α controls the age term, which favors newer strategies to keep diversity and avoid premature convergence.
- **Strategy selection:** during online execution, strategies are drawn by Boltzmann sampling over fitness with a temperature τ, which trades off exploration against exploitation. Only strategies tagged for the current domain(s) are eligible.
- **Evaluation:** experiments use MultiWOZ and SGD with simulated users. Human evaluation is reported to confirm that the evolved strategies are safe, interpretable and useful.
- **Limitations:** per-turn peer critique makes inference latency grow with the number of agents. The offline evolution loop adds compute cost. Reliance on simulation leaves real-world deployment untested, so the authors see batch or latency-tolerant settings as the best fit.
- **Code:** released at github.com/carsonz/DarwinTOD.

*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

- [Shanghai Jiao Tong University](https://md.research.elamur.ai/free/en/universities/shanghai-jiao-tong-university/) — University, Shanghai, China · 212 works · 19,544 citations
- [Chinese Academy of Sciences](https://md.research.elamur.ai/free/en/government/chinese-academy-of-sciences/) — Government body, Beijing, China · 246 works · 24,951 citations
- [The University of Sydney](https://md.research.elamur.ai/free/en/universities/the-university-of-sydney/) — University, Sydney, Australia · 49 works · 68,350 citations
- [Czech Academy of Sciences](https://md.research.elamur.ai/free/en/government/czech-academy-of-sciences/) — Government body, Prague, Czechia · 3 works · 39 citations
- [Beijing Institute of Graphic Communication](https://md.research.elamur.ai/free/en/universities/beijing-institute-of-graphic-communication/) — University, Beijing, China · 3 works

Also on the byline: [Tianjin University of Finance and Economics](https://ror.org/05ev1jb90).

📚 Shelves (4): [🖥️ 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

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🗨️ Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · 654 works

## Cite

Cite as: Shuyu Zhang, Yujie Liu, Xinru Wang, Cheng Zhang, Yanmin Zhu and Bin Li. 2026. DarwinTOD: LLM-Driven Lifelong Self-evolution for Task-oriented Dialog Systems. 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{zhang2026darwintod,
  title = {{DarwinTOD: LLM-Driven Lifelong Self-evolution for Task-oriented Dialog Systems}},
  author = {Shuyu Zhang, Yujie Liu, Xinru Wang, Cheng Zhang, Yanmin Zhu and Bin Li},
  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.2050},
  url = {https://doi.org/10.18653/v1/2026.acl-long.2050}
}
```

- [← previous DART: Mitigating Harm Drift in Difference-Aware LLMs via Distill-Audit-Repair Training](https://md.research.elamur.ai/free/en/papers/pan-2026-dart-mitigating-harm-drift-difference/)
- [next → Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards](https://md.research.elamur.ai/free/en/papers/niu-2026-dashboard2code-evaluating-multimodal-models-reconstructing/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/29/)

## References

56 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Mihail Eric, Rahul Goel, Shachi Paul, Abhishek Sethi, Sanchit Agarwal, Shuyang Gao, Adarsh Kumar, Anuj Goyal, Peter Ku, and Dilek Hakkani-Tur. 2020. MultiWOZ 2.1: A consolidated multi-domain dialogue dataset with state corrections and state tracking baselines. In Proceedings of the Twelfth Language… [find](https://scholar.google.com/scholar?q=Mihail+Eric%2C+Rahul+Goel%2C+Shachi+Paul%2C+Abhishek+Sethi%2C+Sanchit+Agarwal%2C+Shuyang+Gao%2C+Adarsh+Kumar%2C+Anuj+Goyal%2C+Peter+Ku%2C+and+Dilek+Hakkani%2DTur%2E+2020%2E+MultiWOZ+2%2E1%3A+A+consolidated+multi%2Ddomain+dialogue+)
2. Eshaan Agarwal, Raghav Magazine, Joykirat Singh, Vivek Dani, Tanuja Ganu, and Akshay Nambi. 2025. PromptWizard: Optimizing prompts via task-aware, feedback-driven self-evolution. In Findings of the Association for Computational Linguistics: ACL 2025, pages 19974–20003, Vienna, Austria. Association f… [find](https://scholar.google.com/scholar?q=Eshaan+Agarwal%2C+Raghav+Magazine%2C+Joykirat+Singh%2C+Vivek+Dani%2C+Tanuja+Ganu%2C+and+Akshay+Nambi%2E+2025%2E+PromptWizard%3A+Optimizing+prompts+via+task%2Daware%2C+feedback%2Ddriven+self%2Devolution%2E+In+Findings+of+the+As)
3. 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)
4. Namo Bang, Jeehyun Lee, and Myoung-Wan Koo. 2023. Task-optimized adapters for an end-to-end task-oriented dialogue system. In Findings of the Association for Computational Linguistics: ACL 2023, pages 7355–7369, Toronto, Canada. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Namo+Bang%2C+Jeehyun+Lee%2C+and+Myoung%2DWan+Koo%2E+2023%2E+Task%2Doptimized+adapters+for+an+end%2Dto%2Dend+task%2Doriented+dialogue+system%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2023%2C+pages)
5. Chrisantha Fernando, Dylan Banarse, Henryk Michalewski, Simon Osindero, and Tim Rocktäschel. 2024. [Promptbreeder: self-referential self-improvement via prompt evolution](https://md.research.elamur.ai/free/en/papers/fernando-2023-promptbreeder-self-referential-self-improvement/). In Proceedings of the 41st International Conference on Machine Learning, ICML’24. JMLR.org.
6. Paweł Budzianowski, Tsung-Hsien Wen, Bo-Hsiang Tseng, Iñigo Casanueva, Stefan Ultes, Osman Ramadan, and Milica Gašić. 2018. [MultiWOZ - a large-scale multi-domain Wizard-of-Oz dataset for task-oriented dialogue modelling](https://md.research.elamur.ai/free/en/papers/budzianowski-2018-multiwoz-large-scale-multi-domain/). In Proceedings of the 2018 Conference on Empirical Methods in Natural Language…
7. Haoyu Gao, Ting-En Lin, Hangyu Li, Min Yang, Yuchuan Wu, Wentao Ma, Fei Huang, and Yongbin Li. 2024a. Self-explanation prompting improves dialogue understanding in large language models. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Ev… [find](https://scholar.google.com/scholar?q=Haoyu+Gao%2C+Ting%2DEn+Lin%2C+Hangyu+Li%2C+Min+Yang%2C+Yuchuan+Wu%2C+Wentao+Ma%2C+Fei+Huang%2C+and+Yongbin+Li%2E+2024a%2E+Self%2Dexplanation+prompting+improves+dialogue+understanding+in+large+language+models%2E+In+Proceeding)
8. Weize Chen, Jiarui Yuan, Chen Qian, Cheng Yang, Zhiyuan Liu, and Maosong Sun. 2025. Optima: Optimizing effectiveness and efficiency for LLM-based multi-agent system. In Findings of the Association for Computational Linguistics: ACL 2025, pages 11534–11557, Vienna, Austria. Association for Computatio… [find](https://scholar.google.com/scholar?q=Weize+Chen%2C+Jiarui+Yuan%2C+Chen+Qian%2C+Cheng+Yang%2C+Zhiyuan+Liu%2C+and+Maosong+Sun%2E+2025%2E+Optima%3A+Optimizing+effectiveness+and+efficiency+for+LLM%2Dbased+multi%2Dagent+system%2E+In+Findings+of+the+Association+for)
9. Jinglong Gao, Xiao Ding, Yiming Cui, Jianbai Zhao, Hepeng Wang, Ting Liu, and Bing Qin. 2024b. Self-evolving GPT: A lifelong autonomous experiential learner. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6385–6432, Bangkok,… [find](https://scholar.google.com/scholar?q=Jinglong+Gao%2C+Xiao+Ding%2C+Yiming+Cui%2C+Jianbai+Zhao%2C+Hepeng+Wang%2C+Ting+Liu%2C+and+Bing+Qin%2E+2024b%2E+Self%2Devolving+GPT%3A+A+lifelong+autonomous+experiential+learner%2E+In+Proceedings+of+the+62nd+Annual+Meeting+)
10. Yi Cheng, Wenge Liu, Jian Wang, Chak Tou Leong, Yi Ouyang, Wenjie Li, Xian Wu, and Yefeng Zheng. 2024. Cooper: Coordinating specialized agents Yeseul Gong, Heeseon Kim, Seokju Hwang, Donghyun Kim, and Kyong-Ho Lee. 2025. Multi-domain dialogue state tracking via dual dynamic graph with hierarchical s… [find](https://scholar.google.com/scholar?q=Yi+Cheng%2C+Wenge+Liu%2C+Jian+Wang%2C+Chak+Tou+Leong%2C+Yi+Ouyang%2C+Wenjie+Li%2C+Xian+Wu%2C+and+Yefeng+Zheng%2E+2024%2E+Cooper%3A+Coordinating+specialized+agents+Yeseul+Gong%2C+Heeseon+Kim%2C+Seokju+Hwang%2C+Donghyun+Kim%2C+and)
11. Qingyan Guo, Rui Wang, Junliang Guo, Bei Li, Kaitao Song, Xu Tan, Guoqing Liu, Jiang Bian, and Yujiu Yang. 2024. Connecting large language models with evolutionary algorithms yields powerful prompt optimizers. In International Conference on Representation Learning, volume 2024, pages 34133–34156. [find](https://scholar.google.com/scholar?q=Qingyan+Guo%2C+Rui+Wang%2C+Junliang+Guo%2C+Bei+Li%2C+Kaitao+Song%2C+Xu+Tan%2C+Guoqing+Liu%2C+Jiang+Bian%2C+and+Yujiu+Yang%2E+2024%2E+Connecting+large+language+models+with+evolutionary+algorithms+yields+powerful+prompt+op)
12. Wanwei He, Yinpei Dai, Yinhe Zheng, Yuchuan Wu, Zheng Cao, Dermot Liu, Peng Jiang, Min Yang, Fei Huang, Luo Si, Jian Sun, and Yongbin Li. 2022. Galaxy: A generative pre-trained model for task-oriented dialog with semi-supervised learning and explicit policy injection. Proceedings of the AAAI Confere… [find](https://scholar.google.com/scholar?q=Wanwei+He%2C+Yinpei+Dai%2C+Yinhe+Zheng%2C+Yuchuan+Wu%2C+Zheng+Cao%2C+Dermot+Liu%2C+Peng+Jiang%2C+Min+Yang%2C+Fei+Huang%2C+Luo+Si%2C+Jian+Sun%2C+and+Yongbin+Li%2E+2022%2E+Galaxy%3A+A+generative+pre%2Dtrained+model+for+task%2Doriented)
13. Ehsan Hosseini-Asl, Bryan McCann, Chien-Sheng Wu, Semih Yavuz, and Richard Socher. 2020. [A simple language model for task-oriented dialogue](https://md.research.elamur.ai/free/en/papers/hosseini-asl-2020-simple-language-model-task-oriented/). In Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20, Red Hook, NY, USA. Curran Associates Inc.
14. Zhenhua Huang, Fancong Li, Juanjuan Yao, and Zong-gan Chen. 2023. Mgcrl: Multi-view graph convolution and multi-agent reinforcement learning for dialogue state tracking. Neural Comput. Appl., 36(9):4829–4846. [find](https://scholar.google.com/scholar?q=Zhenhua+Huang%2C+Fancong+Li%2C+Juanjuan+Yao%2C+and+Zong%2Dgan+Chen%2E+2023%2E+Mgcrl%3A+Multi%2Dview+graph+convolution+and+multi%2Dagent+reinforcement+learning+for+dialogue+state+tracking%2E+Neural+Comput%2E+Appl%2E%2C+36%289%29%3A48)
15. Minji Kim, Joon Yoo, and OkRan Jeong. 2025. Medi-toder: Medical domain-incremental task-oriented dialogue generator using experience replay. Expert Systems, 42(2):e13773. [find](https://scholar.google.com/scholar?q=Minji+Kim%2C+Joon+Yoo%2C+and+OkRan+Jeong%2E+2025%2E+Medi%2Dtoder%3A+Medical+domain%2Dincremental+task%2Doriented+dialogue+generator+using+experience+replay%2E+Expert+Systems%2C+42%282%29%3Ae13773%2E)
16. Brendan King and Jeffrey Flanigan. 2024. Unsupervised end-to-end task-oriented dialogue with LLMs: The power of the noisy channel. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 8283–8300, Miami, Florida, USA. Association for Computational Linguistic… [find](https://scholar.google.com/scholar?q=Brendan+King+and+Jeffrey+Flanigan%2E+2024%2E+Unsupervised+end%2Dto%2Dend+task%2Doriented+dialogue+with+LLMs%3A+The+power+of+the+noisy+channel%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+in+Natural)
17. Zhaojiang Lin, Andrea Madotto, Genta Indra Winata, and Pascale Fung. 2020. [MinTL: Minimalist transfer learning for task-oriented dialogue systems](https://md.research.elamur.ai/free/en/papers/lin-2020-mintl-minimalist-transfer-learning-task/). In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 3391–3405, Online. Association for Computational…
18. Bing Liu and Sahisnu Mazumder. 2021. Lifelong and continual learning dialogue systems: Learning during conversation. Proceedings of the AAAI Conference on Artificial Intelligence, 35(17):15058–15063. [find](https://scholar.google.com/scholar?q=Bing+Liu+and+Sahisnu+Mazumder%2E+2021%2E+Lifelong+and+continual+learning+dialogue+systems%3A+Learning+during+conversation%2E+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+35%2817%29%3A15058%E2%80%9315063%2E)
19. Andrea Madotto, Zhaojiang Lin, Zhenpeng Zhou, Seungwhan Moon, Paul Crook, Bing Liu, Zhou Yu, Eunjoon Cho, Pascale Fung, and Zhiguang Wang. 2021. Continual learning in task-oriented dialogue systems. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7452… [find](https://scholar.google.com/scholar?q=Andrea+Madotto%2C+Zhaojiang+Lin%2C+Zhenpeng+Zhou%2C+Seungwhan+Moon%2C+Paul+Crook%2C+Bing+Liu%2C+Zhou+Yu%2C+Eunjoon+Cho%2C+Pascale+Fung%2C+and+Zhiguang+Wang%2E+2021%2E+Continual+learning+in+task%2Doriented+dialogue+systems%2E+I)
20. Samir W. Mahfoud. 1996. Niching methods for genetic algorithms. Ph.D. thesis, University of Illinois at Urbana-Champaign, USA. UMI Order No. GAX95-43663. [find](https://scholar.google.com/scholar?q=Samir+W%2E+Mahfoud%2E+1996%2E+Niching+methods+for+genetic+algorithms%2E+Ph%2ED%2E+thesis%2C+University+of+Illinois+at+Urbana%2DChampaign%2C+USA%2E+UMI+Order+No%2E+GAX95%2D43663%2E)
21. Sahisnu Mazumder and Bing Liu. 2022. Continual learning dialogue systems - learning during conversation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’22, page 3429–3432, New York, NY, USA. Association for Computing Machine… [find](https://scholar.google.com/scholar?q=Sahisnu+Mazumder+and+Bing+Liu%2E+2022%2E+Continual+learning+dialogue+systems+%2D+learning+during+conversation%2E+In+Proceedings+of+the+45th+International+ACM+SIGIR+Conference+on+Research+and+Development+in+In)
22. Adib Mosharrof, Muhammad Hasan Maqbool, and A.B. Siddique. 2023. Zero-shot generalizable end-to-end task-oriented dialog system using context summarization and domain schema. The International FLAIRS Conference Proceedings, 36. [find](https://scholar.google.com/scholar?q=Adib+Mosharrof%2C+Muhammad+Hasan+Maqbool%2C+and+A%2EB%2E+Siddique%2E+2023%2E+Zero%2Dshot+generalizable+end%2Dto%2Dend+task%2Doriented+dialog+system+using+context+summarization+and+domain+schema%2E+The+International+FLAIRS+)
23. Efe Precious Onakpojeruo, Berna Uzun, Leena R. David, Ilker Ozsahin, and Dilber Uzun Ozsahin. 2024. Selection techniques in genetic algorithm. In 2024 17th International Conference on Development in eSystem Engineering (DeSE), pages 411–416. [find](https://scholar.google.com/scholar?q=Efe+Precious+Onakpojeruo%2C+Berna+Uzun%2C+Leena+R%2E+David%2C+Ilker+Ozsahin%2C+and+Dilber+Uzun+Ozsahin%2E+2024%2E+Selection+techniques+in+genetic+algorithm%2E+In+2024+17th+International+Conference+on+Development+in+e)
24. Robert Lange, Yingtao Tian, and Yujin Tang. 2024. Large language models as evolution strategies. In Proceedings of the Genetic and Evolutionary Computation Conference Companion, GECCO ’24 Companion, page 579–582, New York, NY, USA. Association for Computing Machinery. [find](https://scholar.google.com/scholar?q=Robert+Lange%2C+Yingtao+Tian%2C+and+Yujin+Tang%2E+2024%2E+Large+language+models+as+evolution+strategies%2E+In+Proceedings+of+the+Genetic+and+Evolutionary+Computation+Conference+Companion%2C+GECCO+%E2%80%9924+Companion%2C+p)
25. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. [Bleu: a method for automatic evaluation of machine translation](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/). In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational…
26. Harrison Lee, Raghav Gupta, Abhinav Rastogi, Yuan Cao, Bin Zhang, and Yonghui Wu. 2022. Sgd-x: A benchmark for robust generalization in schema-guided dialogue systems. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 10938–10946. [find](https://scholar.google.com/scholar?q=Harrison+Lee%2C+Raghav+Gupta%2C+Abhinav+Rastogi%2C+Yuan+Cao%2C+Bin+Zhang%2C+and+Yonghui+Wu%2E+2022%2E+Sgd%2Dx%3A+A+benchmark+for+robust+generalization+in+schema%2Dguided+dialogue+systems%2E+In+Proceedings+of+the+AAAI+Confe)
27. Baolin Peng, Chunyuan Li, Jinchao Li, Shahin Shayandeh, Lars Liden, and Jianfeng Gao. 2021. Soloist: Building task bots at scale with transfer learning and machine teaching. Transactions of the Association for Computational Linguistics, 9:807–824. [find](https://scholar.google.com/scholar?q=Baolin+Peng%2C+Chunyuan+Li%2C+Jinchao+Li%2C+Shahin+Shayandeh%2C+Lars+Liden%2C+and+Jianfeng+Gao%2E+2021%2E+Soloist%3A+Building+task+bots+at+scale+with+transfer+learning+and+machine+teaching%2E+Transactions+of+the+Associ)
28. Pan Li, Qingwen Yang, Sai Xu, Xuejing Li, Zhi Li, Chen Wang, Yanyi Liu, Tiezheng Guo, Jiawei Tang, and Yingyou Wen. 2025. Adaptive-tod: An llm-driven and adaptive agent for diverse interaction modes. Neurocomputing, 652:130991. [find](https://scholar.google.com/scholar?q=Pan+Li%2C+Qingwen+Yang%2C+Sai+Xu%2C+Xuejing+Li%2C+Zhi+Li%2C+Chen+Wang%2C+Yanyi+Liu%2C+Tiezheng+Guo%2C+Jiawei+Tang%2C+and+Yingyou+Wen%2E+2025%2E+Adaptive%2Dtod%3A+An+llm%2Ddriven+and+adaptive+agent+for+diverse+interaction+modes%2E+)
29. Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, and Pranav Khaitan. 2020. [Towards scalable multi-domain conversational agents: The schema-guided dialogue dataset](https://md.research.elamur.ai/free/en/papers/rastogi-2019-towards-scalable-multi-domain-conversational/). Proceedings of the AAAI Conference on Artificial Intelligence, 34(05):8689–8696.
30. Anton Razzhigaev, Maxim Kurkin, Elizaveta Goncharova, Irina Abdullaeva, Anastasia Lysenko, Alexander Panchenko, Andrey Kuznetsov, and Denis Dimitrov. 2024. OmniDialog: A multimodal benchmark for generalization across text, visual, and audio modalities. In Proceedings of the 2nd GenBench Workshop on… [find](https://scholar.google.com/scholar?q=Anton+Razzhigaev%2C+Maxim+Kurkin%2C+Elizaveta+Goncharova%2C+Irina+Abdullaeva%2C+Anastasia+Lysenko%2C+Alexander+Panchenko%2C+Andrey+Kuznetsov%2C+and+Denis+Dimitrov%2E+2024%2E+OmniDialog%3A+A+multimodal+benchmark+for+gener)
31. Jeonghoon Shim, Gyuhyeon Seo, Cheongsu Lim, and Yohan Jo. 2025. Tooldial: Multi-turn dialogue generation method for tool-augmented language models. In International Conference on Representation Learning, volume 2025, pages 70414–70439. [find](https://scholar.google.com/scholar?q=Jeonghoon+Shim%2C+Gyuhyeon+Seo%2C+Cheongsu+Lim%2C+and+Yohan+Jo%2E+2025%2E+Tooldial%3A+Multi%2Dturn+dialogue+generation+method+for+tool%2Daugmented+language+models%2E+In+International+Conference+on+Representation+Learni)
32. Haoyang Su, Renqi Chen, Shixiang Tang, Zhenfei Yin, Xinzhe Zheng, Jinzhe Li, Biqing Qi, Qi Wu, Hui Li, Wanli Ouyang, Philip Torr, Bowen Zhou, and Nanqing Dong. 2025. [Many heads are better than one: Improved scientific idea generation by a LLM-based multi-agent system](https://md.research.elamur.ai/free/en/papers/su-2025-many-heads-are-better-than/). In Proceedings of the 63rd Annu…
33. Yixuan Su, Lei Shu, Elman Mansimov, Arshit Gupta, Deng Cai, Yi-An Lai, and Yi Zhang. 2022. Multi-task pre-training for plug-and-play task-oriented dialogue system. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4661–4676, Dub… [find](https://scholar.google.com/scholar?q=Yixuan+Su%2C+Lei+Shu%2C+Elman+Mansimov%2C+Arshit+Gupta%2C+Deng+Cai%2C+Yi%2DAn+Lai%2C+and+Yi+Zhang%2E+2022%2E+Multi%2Dtask+pre%2Dtraining+for+plug%2Dand%2Dplay+task%2Doriented+dialogue+system%2E+In+Proceedings+of+the+60th+Annual+Me)
34. Haipeng Sun, Junwei Bao, Youzheng Wu, and Xiaodong He. 2022. BORT: Back and denoising reconstruction for end-to-end task-oriented dialog. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 2156–2170, Seattle, United States. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Haipeng+Sun%2C+Junwei+Bao%2C+Youzheng+Wu%2C+and+Xiaodong+He%2E+2022%2E+BORT%3A+Back+and+denoising+reconstruction+for+end%2Dto%2Dend+task%2Doriented+dialog%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+)
35. Zhen Tan, Jun Yan, I-Hung Hsu, Rujun Han, Zifeng Wang, Long Le, Yiwen Song, Yanfei Chen, Hamid Palangi, George Lee, Anand Rajan Iyer, Tianlong Chen, Huan Liu, Chen-Yu Lee, and Tomas Pfister. 2025. [In prospect and retrospect: Reflective memory management for long-term personalized dialogue agents](https://md.research.elamur.ai/free/en/papers/tan-2025-prospect-retrospect-reflective-memory-management/). In…
36. Zhengwei Tao, Ting-En Lin, Xiancai Chen, Hangyu Li, Yuchuan Wu, Yongbin Li, Zhi Jin, Fei Huang, Dacheng Tao, and Jingren Zhou. 2024. A survey on self-evolution of large language models. Preprint, arXiv:2404.14387. — [arXiv:2404.14387](https://arxiv.org/abs/2404.14387) [find](https://scholar.google.com/scholar?q=Zhengwei+Tao%2C+Ting%2DEn+Lin%2C+Xiancai+Chen%2C+Hangyu+Li%2C+Yuchuan+Wu%2C+Yongbin+Li%2C+Zhi+Jin%2C+Fei+Huang%2C+Dacheng+Tao%2C+and+Jingren+Zhou%2E+2024%2E+A+survey+on+self%2Devolution+of+large+language+models%2E+Preprint%2C+arXi)
37. Graham Todd, Alexander G. Padula, Matthew Stephenson, Éric Piette, Dennis J.N.J. Soemers, and Julian Togelius. 2024. Gavel: generating games via evolution and language models. In Proceedings of the 38th International Conference on Neural Information Processing Systems, NIPS ’24, Red Hook, NY, USA. C… [find](https://scholar.google.com/scholar?q=Graham+Todd%2C+Alexander+G%2E+Padula%2C+Matthew+Stephenson%2C+%C3%89ric+Piette%2C+Dennis+J%2EN%2EJ%2E+Soemers%2C+and+Julian+Togelius%2E+2024%2E+Gavel%3A+generating+games+via+evolution+and+language+models%2E+In+Proceedings+of+the+38)
38. Laurens van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-sne. Journal of Machine Learning Research, 9(86):2579–2605. [find](https://scholar.google.com/scholar?q=Laurens+van+der+Maaten+and+Geoffrey+Hinton%2E+2008%2E+Visualizing+data+using+t%2Dsne%2E+Journal+of+Machine+Learning+Research%2C+9%2886%29%3A2579%E2%80%932605%2E)
39. Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel Bowman. 2018. [GLUE: A multi-task benchmark and analysis platform for natural language understanding](https://md.research.elamur.ai/free/en/papers/wang-2018-glue-multi-task-benchmark-analysis/). In Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, pages 353–355,…
40. Chien-Sheng Wu, Andrea Madotto, Ehsan Hosseini-Asl, Caiming Xiong, Richard Socher, and Pascale Fung. 2019. [Transferable multi-domain state generator for task-oriented dialogue systems](https://md.research.elamur.ai/free/en/papers/wu-2019-transferable-multi-domain-state-generator/). In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 808–819, Florence…
41. Xingyu Wu, Sheng-Hao Wu, Jibin Wu, Liang Feng, and Kay Chen Tan. 2025. Evolutionary computation in the era of large language model: Survey and roadmap. IEEE Transactions on Evolutionary Computation, 29(2):534–554. [find](https://scholar.google.com/scholar?q=Xingyu+Wu%2C+Sheng%2DHao+Wu%2C+Jibin+Wu%2C+Liang+Feng%2C+and+Kay+Chen+Tan%2E+2025%2E+Evolutionary+computation+in+the+era+of+large+language+model%3A+Survey+and+roadmap%2E+IEEE+Transactions+on+Evolutionary+Computation%2C+2)
42. Heng-Da Xu, Xian-Ling Mao, Fanshu Sun, Tian-Yi Che, Chun Xu, and Heyan Huang. 2025a. Agenttod: A task-oriented dialogue agent with a flexible and adaptive api calling paradigm. ACM Trans. Inf. Syst., 43(5). [find](https://scholar.google.com/scholar?q=Heng%2DDa+Xu%2C+Xian%2DLing+Mao%2C+Fanshu+Sun%2C+Tian%2DYi+Che%2C+Chun+Xu%2C+and+Heyan+Huang%2E+2025a%2E+Agenttod%3A+A+task%2Doriented+dialogue+agent+with+a+flexible+and+adaptive+api+calling+paradigm%2E+ACM+Trans%2E+Inf%2E+Syst%2E%2C+)
43. Heng-Da Xu, Xian-Ling Mao, Fanshu Sun, Tian-Yi Che, Chun Xu, and Heyan Huang. 2025b. Agenttod: A task-oriented dialogue agent with a flexible and adaptive api calling paradigm. ACM Transactions on Information Systems, 43(5):1–32. [find](https://scholar.google.com/scholar?q=Heng%2DDa+Xu%2C+Xian%2DLing+Mao%2C+Fanshu+Sun%2C+Tian%2DYi+Che%2C+Chun+Xu%2C+and+Heyan+Huang%2E+2025b%2E+Agenttod%3A+A+task%2Doriented+dialogue+agent+with+a+flexible+and+adaptive+api+calling+paradigm%2E+ACM+Transactions+on+Inf)
44. Heng-Da Xu, Xian-Ling Mao, Puhai Yang, Fanshu Sun, and Heyan Huang. 2024. Rethinking task-oriented dialogue systems: From complex modularity to zero-shot autonomous agent. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2748–… [find](https://scholar.google.com/scholar?q=Heng%2DDa+Xu%2C+Xian%2DLing+Mao%2C+Puhai+Yang%2C+Fanshu+Sun%2C+and+Heyan+Huang%2E+2024%2E+Rethinking+task%2Doriented+dialogue+systems%3A+From+complex+modularity+to+zero%2Dshot+autonomous+agent%2E+In+Proceedings+of+the+62nd+A)
45. Qiancheng Xu, Min Yang, and Ruifeng Xu. 2023. Balanced meta learning and diverse sampling for lifelong task-oriented dialogue systems. Proceedings of the AAAI Conference on Artificial Intelligence, 37(11):13843–13852. [find](https://scholar.google.com/scholar?q=Qiancheng+Xu%2C+Min+Yang%2C+and+Ruifeng+Xu%2E+2023%2E+Balanced+meta+learning+and+diverse+sampling+for+lifelong+task%2Doriented+dialogue+systems%2E+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+37)
46. Yunyi Yang, Yunhao Li, and Xiaojun Quan. 2021. Ubar: Towards fully end-to-end task-oriented dialog system with gpt-2. Proceedings of the AAAI Conference on Artificial Intelligence, 35(16):14230–14238. [find](https://scholar.google.com/scholar?q=Yunyi+Yang%2C+Yunhao+Li%2C+and+Xiaojun+Quan%2E+2021%2E+Ubar%3A+Towards+fully+end%2Dto%2Dend+task%2Doriented+dialog+system+with+gpt%2D2%2E+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+35%2816%29%3A14230%E2%80%9314238%2E)
47. Fanghua Ye, Jarana Manotumruksa, and Emine Yilmaz. 2022. [MultiWOZ 2.4: A multi-domain task-oriented dialogue dataset with essential annotation corrections to improve state tracking evaluation](https://md.research.elamur.ai/free/en/papers/ye-2022-multiwoz-2-4-multi-domain/). In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 35…
48. Xiaoxue Zang, Abhinav Rastogi, Srinivas Sunkara, Raghav Gupta, Jianguo Zhang, and Jindong Chen. 2020. [MultiWOZ 2.2 : A dialogue dataset with additional annotation corrections and state tracking baselines](https://md.research.elamur.ai/free/en/papers/zang-2020-multiwoz-2-2-dialogue-dataset/). In Proceedings of the 2nd Workshop on Natural Language Processing for Conversational AI, pages…
49. Min Zeng, Haiqin Yang, Xi Chen, and Yike Guo. 2025. Task-wrapped continual learning in task-oriented dialogue systems. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 3173–3183, Albuquerque, New Mexico. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Min+Zeng%2C+Haiqin+Yang%2C+Xi+Chen%2C+and+Yike+Guo%2E+2025%2E+Task%2Dwrapped+continual+learning+in+task%2Doriented+dialogue+systems%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+NAACL+2025%2C+pages+3)
50. Yunpeng Zhai, Shuchang Tao, Cheng Chen, Anni Zou, Ziqian Chen, Qingxu Fu, Shinji Mai, Li Yu, Jiaji Deng, Zouying Cao, Zhaoyang Liu, Bolin Ding, and Jingren Zhou. 2025. Agentevolver: Towards efficient self-evolving agent system. Preprint, arXiv:2511.10395. — [arXiv:2511.10395](https://arxiv.org/abs/2511.10395) [find](https://scholar.google.com/scholar?q=Yunpeng+Zhai%2C+Shuchang+Tao%2C+Cheng+Chen%2C+Anni+Zou%2C+Ziqian+Chen%2C+Qingxu+Fu%2C+Shinji+Mai%2C+Li+Yu%2C+Jiaji+Deng%2C+Zouying+Cao%2C+Zhaoyang+Liu%2C+Bolin+Ding%2C+and+Jingren+Zhou%2E+2025%2E+Agentevolver%3A+Towards+efficient+)
51. Guibin Zhang, Muxin Fu, and Shuicheng Yan. 2025a. Memgen: Weaving generative latent memory for self-evolving agents. Preprint, arXiv:2509.24704. — [arXiv:2509.24704](https://arxiv.org/abs/2509.24704) [find](https://scholar.google.com/scholar?q=Guibin+Zhang%2C+Muxin+Fu%2C+and+Shuicheng+Yan%2E+2025a%2E+Memgen%3A+Weaving+generative+latent+memory+for+self%2Devolving+agents%2E+Preprint%2C+arXiv%3A2509%2E24704%2E+%E2%80%94+%5BarXiv%3A2509%2E24704%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2509%2E24704%29)
52. Jian Zhang, Zhangqi Wang, Haiping Zhu, Kangda Cheng, Kai He, Bo Li, Qika Lin, Jun Liu, and Erik Cambria. 2025b. Mars: Multi-agent adaptive reasoning with socratic guidance for automated prompt optimization. Preprint, arXiv:2503.16874. — [arXiv:2503.16874](https://arxiv.org/abs/2503.16874) [find](https://scholar.google.com/scholar?q=Jian+Zhang%2C+Zhangqi+Wang%2C+Haiping+Zhu%2C+Kangda+Cheng%2C+Kai+He%2C+Bo+Li%2C+Qika+Lin%2C+Jun+Liu%2C+and+Erik+Cambria%2E+2025b%2E+Mars%3A+Multi%2Dagent+adaptive+reasoning+with+socratic+guidance+for+automated+prompt+optimiz)
53. Yichi Zhang, Zhijian Ou, and Zhou Yu. 2020. Task-oriented dialog systems that consider multiple appropriate responses under the same context. Proceedings of the AAAI Conference on Artificial Intelligence, 34(05):9604–9611. [find](https://scholar.google.com/scholar?q=Yichi+Zhang%2C+Zhijian+Ou%2C+and+Zhou+Yu%2E+2020%2E+Task%2Doriented+dialog+systems+that+consider+multiple+appropriate+responses+under+the+same+context%2E+Proceedings+of+the+AAAI+Conference+on+Artificial+Intellige)
54. Yisong Zhang and Guoxing Yi. 2025. Laos: Large language model-driven adaptive operator selection for evolutionary algorithms. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’25, page 517–526, New York, NY, USA. Association for Computing Machinery. [find](https://scholar.google.com/scholar?q=Yisong+Zhang+and+Guoxing+Yi%2E+2025%2E+Laos%3A+Large+language+model%2Ddriven+adaptive+operator+selection+for+evolutionary+algorithms%2E+In+Proceedings+of+the+Genetic+and+Evolutionary+Computation+Conference%2C+GEC)
55. Yangyang Zhao, Ben Niu, Libo Qin, and Shihan Wang. 2025. An efficient task-oriented dialogue policy: Evolutionary reinforcement learning injected by elite individuals. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3429–3442,… [find](https://scholar.google.com/scholar?q=Yangyang+Zhao%2C+Ben+Niu%2C+Libo+Qin%2C+and+Shihan+Wang%2E+2025%2E+An+efficient+task%2Doriented+dialogue+policy%3A+Evolutionary+reinforcement+learning+injected+by+elite+individuals%2E+In+Proceedings+of+the+63rd+Annua)
56. Yingxiu Zhao, Yinhe Zheng, Zhiliang Tian, Chang Gao, Jian Sun, and Nevin L. Zhang. 2022. Prompt conditioned VAE: Enhancing generative replay for lifelong learning in task-oriented dialogue. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 11153–11169,… [find](https://scholar.google.com/scholar?q=Yingxiu+Zhao%2C+Yinhe+Zheng%2C+Zhiliang+Tian%2C+Chang+Gao%2C+Jian+Sun%2C+and+Nevin+L%2E+Zhang%2E+2022%2E+Prompt+conditioned+VAE%3A+Enhancing+generative+replay+for+lifelong+learning+in+task%2Doriented+dialogue%2E+In+Proceed)

*Updated 2 November 2026*