---
title: "GraphMind — paper by Shin et al. (2026) — summary and ideas"
description: "LLM agents in interactive environments usually reason over only a short window of recent steps. 34 references."
authors: ["Sunguk Shin", "Hayeong Lee", "Jun Ho Seo", "Jinho Lee", "Myunsoo Kim", "Minsuk Chang", "Byung-Jun Lee"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1651/"
doi: "10.18653/v1/2026.findings-acl.1651"
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/shin-2026-graphmind-llms-as-dynamic-knowledge/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/shin-2026-graphmind-llms-as-dynamic-knowledge/)

[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/31/) / GraphMind: LLMs as Dynamic Knowledge Builders for Sequential Decision-Making

# GraphMind: LLMs as Dynamic Knowledge Builders for Sequential Decision-Making

**Authors:** [Sunguk Shin](https://md.research.elamur.ai/free/en/authors/sunguk-shin/), Hayeong Lee, [Jun Ho Seo](https://md.research.elamur.ai/free/en/authors/jun-ho-seo/), Jinho Lee, Myunsoo Kim, Minsuk Chang and [Byung-Jun Lee](https://md.research.elamur.ai/free/en/authors/byung-jun-lee/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/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.findings-acl.1651/) · [PDF](https://aclanthology.org/2026.findings-acl.1651.pdf) · [DOI 10.18653/v1/2026.findings-acl.1651](https://doi.org/10.18653/v1/2026.findings-acl.1651) · [OpenAlex](https://openalex.org/W7166803603)

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

## Summary
LLM agents in interactive environments usually reason over only a short window of recent steps. This approach fails when tasks are long and the world is only partly visible. Retrieval-augmented methods such as GraphRAG rely on fixed, pre-built knowledge, so they cannot absorb facts that emerge during interaction. GraphMind addresses this by having the LLM build its own knowledge graph while it explores. Nodes stand for rooms and objects, and edges capture spatial and semantic relations. Four LLM-driven modules handle the graph:
- one detects room changes,
- one parses observations,
- one localizes the agent,
- one merges new attributes.

Rather than inserting the whole memory into every prompt, the agent queries the graph through tools on demand, including a shortest-path search. Plans are written in a domain-specific language (DSL) that a low-level actor executes, with completion checks and plan revision in a loop. On a harder variant of the BabyAI gridworld, adding the graph memory improved task success, planning efficiency and robustness to restricted visibility. The work suggests that self-built structured memory can make LLMs stronger long-horizon decision-makers.

## Key points
- The graph uses four relation types: is_connected_with, contains, is_located_to and are_apart. The last one distinguishes otherwise identical objects by relative distance.
- Evaluation uses BabyAI with four-room and nine-room layouts of 25 cells each. The nine-room (3 × 3) setup matches BossLevel, and the missions are OpenDoor and PutNextTo.
- The environment was modified so that open doors also block line of sight. This intensifies partial observability, and the BabyAI Oracle Solver (BOT) serves as a reference for analysis.
- Retrieval tools return information about specified and adjacent rooms plus a shortest path to the nearest target, computed with BFS.
- In the stated limitations, the authors acknowledge that mistakes made while building the KG can propagate and cause plans to fail, for example when a room the agent returns to is recorded as a new node. The method may also be hard to extend to open worlds like Minecraft. The code is released at github.com/ku-dmlab/GraphMind.

*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

- [Google DeepMind](https://md.research.elamur.ai/free/en/companies/google-deepmind/) — Company, London, United Kingdom · 56 works · 78,880 citations
- [Korea University](https://md.research.elamur.ai/free/en/universities/korea-university/) — University, Seoul, South Korea · 26 works · 9,490 citations

📚 Shelves (2): [🖥️ 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

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works

## Cite

Cite as: Sunguk Shin, Hayeong Lee, Jun Ho Seo, Jinho Lee, Myunsoo Kim, Minsuk Chang and Byung-Jun Lee. 2026. GraphMind: LLMs as Dynamic Knowledge Builders for Sequential Decision-Making. Findings of the Association for Computational Linguistics: ACL 2026.

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

```bibtex
@inproceedings{shin2026graphmind,
  title = {{GraphMind: LLMs as Dynamic Knowledge Builders for Sequential Decision-Making}},
  author = {Sunguk Shin, Hayeong Lee, Jun Ho Seo, Jinho Lee, Myunsoo Kim, Minsuk Chang and Byung-Jun Lee},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1651},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1651}
}
```

- [← previous GraphLoRA: Structure-Aware Low-Rank Adaptation for Large Language Model Recommendation](https://md.research.elamur.ai/free/en/papers/mu-2026-graphlora-structure-aware-low-rank/)
- [next → GraphRAG-Rad: Concept-Aware Radiology Report Generation via Latent Visual-Semantic Retrieval](https://md.research.elamur.ai/free/en/papers/safari-2026-graphrag-rad-concept-aware-radiology/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/31/)

## References

34 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Petr Anokhin, Nikita Semenov, Artyom Sorokin, Dmitry Evseev, Andrey Kravchenko, Mikhail Burtsev, and Evgeny Burnaev. 2024. Arigraph: Learning knowledge graph world models with episodic memory for llm agents. arXiv preprint arXiv:2407.04363. — [arXiv:2407.04363](https://arxiv.org/abs/2407.04363) [find](https://scholar.google.com/scholar?q=Petr+Anokhin%2C+Nikita+Semenov%2C+Artyom+Sorokin%2C+Dmitry+Evseev%2C+Andrey+Kravchenko%2C+Mikhail+Burtsev%2C+and+Evgeny+Burnaev%2E+2024%2E+Arigraph%3A+Learning+knowledge+graph+world+models+with+episodic+memory+for+llm+)
2. Anthropic. 2025. Introducing claude 4. [https://www.anthropic.com/news/claude-4](https://www.anthropic.com/news/claude-4). [↗](https://www.anthropic.com/news/claude-4)
3. Shraddha Barke, Emmanuel Anaya Gonzalez, Saketh Ram Kasibatla, Taylor Berg-Kirkpatrick, and Nadia Polikarpova. 2024. Hysynth: Context-free llm approximation for guiding program synthesis. Advances in Neural Information Processing Systems, 37:15612–15645. [find](https://scholar.google.com/scholar?q=Shraddha+Barke%2C+Emmanuel+Anaya+Gonzalez%2C+Saketh+Ram+Kasibatla%2C+Taylor+Berg%2DKirkpatrick%2C+and+Nadia+Polikarpova%2E+2024%2E+Hysynth%3A+Context%2Dfree+llm+approximation+for+guiding+program+synthesis%2E+Advances+in+)
4. Thomas Carta, Clément Romac, Thomas Wolf, Sylvain Lamprier, Olivier Sigaud, and Pierre-Yves Oudeyer. 2023. Grounding large language models in interactive environments with online reinforcement learning. In International Conference on Machine Learning, pages 3676–3713. PMLR. [find](https://scholar.google.com/scholar?q=Thomas+Carta%2C+Cl%C3%A9ment+Romac%2C+Thomas+Wolf%2C+Sylvain+Lamprier%2C+Olivier+Sigaud%2C+and+Pierre%2DYves+Oudeyer%2E+2023%2E+Grounding+large+language+models+in+interactive+environments+with+online+reinforcement+learnin)
5. Maxime Chevalier-Boisvert, Dzmitry Bahdanau, Salem Lahlou, Lucas Willems, Chitwan Saharia, Thien Huu Nguyen, and Yoshua Bengio. 2019. Babyai: First steps towards grounded language learning with a human in the loop. In International Conference on Learning Representations, volume 105. New Orleans, LA. [find](https://scholar.google.com/scholar?q=Maxime+Chevalier%2DBoisvert%2C+Dzmitry+Bahdanau%2C+Salem+Lahlou%2C+Lucas+Willems%2C+Chitwan+Saharia%2C+Thien+Huu+Nguyen%2C+and+Yoshua+Bengio%2E+2019%2E+Babyai%3A+First+steps+towards+grounded+language+learning+with+a+huma)
6. Francois Chollet, Mike Knoop, Gregory Kamradt, and Bryan Landers. 2024. Arc prize 2024: Technical report. arXiv preprint arXiv:2412.04604. [find](https://scholar.google.com/scholar?q=Francois+Chollet%2C+Mike+Knoop%2C+Gregory+Kamradt%2C+and+Bryan+Landers%2E+2024%2E+Arc+prize+2024%3A+Technical+report%2E+arXiv+preprint+arXiv%3A2412%2E04604%2E)
7. Nathan Cloos, Meagan Jens, Michelangelo Naim, Yen-Ling Kuo, Ignacio Cases, Andrei Barbu, and Christopher J Cueva. 2024. Baba is ai: Break the rules to beat the benchmark. arXiv preprint arXiv:2407.13729. — [arXiv:2407.13729](https://arxiv.org/abs/2407.13729) [find](https://scholar.google.com/scholar?q=Nathan+Cloos%2C+Meagan+Jens%2C+Michelangelo+Naim%2C+Yen%2DLing+Kuo%2C+Ignacio+Cases%2C+Andrei+Barbu%2C+and+Christopher+J+Cueva%2E+2024%2E+Baba+is+ai%3A+Break+the+rules+to+beat+the+benchmark%2E+arXiv+preprint+arXiv%3A2407%2E137)
8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, and 1 others. 2025. [Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities](https://md.research.elamur.ai/free/en/papers/comanici-2025-gemini-2-5-pushing-frontier/). arX…
9. Marc-Alexandre Côté, Akos Kádár, Xingdi Yuan, Ben Kybartas, Tavian Barnes, Emery Fine, James Moore, Matthew Hausknecht, Layla El Asri, Mahmoud Adada, and 1 others. 2018. [Textworld: A learning environment for text-based games](https://md.research.elamur.ai/free/en/papers/cote-2018-textworld-learning-environment-text-based/). In Workshop on Computer Games, pages 41–75. Springer.
10. Haoyu Han, Yu Wang, Harry Shomer, Kai Guo, Jiayuan Ding, Yongjia Lei, Mahantesh Halappanavar, Ryan A Rossi, Subhabrata Mukherjee, Xianfeng Tang, and 1 others. 2024. Retrieval-augmented generation with graphs (graphrag). arXiv preprint arXiv:2501.00309. [find](https://scholar.google.com/scholar?q=Haoyu+Han%2C+Yu+Wang%2C+Harry+Shomer%2C+Kai+Guo%2C+Jiayuan+Ding%2C+Yongjia+Lei%2C+Mahantesh+Halappanavar%2C+Ryan+A+Rossi%2C+Subhabrata+Mukherjee%2C+Xianfeng+Tang%2C+and+1+others%2E+2024%2E+Retrieval%2Daugmented+generation+with)
11. Zifan He, Yingqi Cao, Zongyue Qin, Neha Prakriya, Yizhou Sun, and Jason Cong. 2025. HMT: Hierarchical memory transformer for efficient long context language processing. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Huma… [find](https://scholar.google.com/scholar?q=Zifan+He%2C+Yingqi+Cao%2C+Zongyue+Qin%2C+Neha+Prakriya%2C+Yizhou+Sun%2C+and+Jason+Cong%2E+2025%2E+HMT%3A+Hierarchical+memory+transformer+for+efficient+long+context+language+processing%2E+In+Proceedings+of+the+2025+Conf)
12. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021. [Measuring mathematical problem solving with the math dataset](https://md.research.elamur.ai/free/en/papers/hendrycks-2021-measuring-mathematical-problem-solving-math/). arXiv preprint arXiv:2103.03874.
13. Aidan Hogan, Eva Blomqvist, Michael Cochez, Claudia d’Amato, Gerard De Melo, Claudio Gutierrez, Sabrina Kirrane, José Emilio Labra Gayo, Roberto Navigli, Sebastian Neumaier, and 1 others. 2021. [find](https://scholar.google.com/scholar?q=Aidan+Hogan%2C+Eva+Blomqvist%2C+Michael+Cochez%2C+Claudia+d%E2%80%99Amato%2C+Gerard+De+Melo%2C+Claudio+Gutierrez%2C+Sabrina+Kirrane%2C+Jos%C3%A9+Emilio+Labra+Gayo%2C+Roberto+Navigli%2C+Sebastian+Neumaier%2C+and+1+others%2E+2021%2E)
14. Knowledge graphs. ACM Computing Surveys (Csur), 54(4):1–37. [find](https://scholar.google.com/scholar?q=Knowledge+graphs%2E+ACM+Computing+Surveys+%28Csur%29%2C+54%284%29%3A1%E2%80%9337%2E)
15. Mengkang Hu, Tianxing Chen, Qiguang Chen, Yao Mu, Wenqi Shao, and Ping Luo. 2025. [HiAgent: Hierarchical working memory management for solving long-horizon agent tasks with large language model](https://md.research.elamur.ai/free/en/papers/hu-2025-hiagent-hierarchical-working-memory-management/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long…
16. Adam Tauman Kalai, Ofir Nachum, Santosh S. Vempala, and Edwin Zhang. 2025. Why language models hallucinate. Preprint, arXiv:2509.04664. — [arXiv:2509.04664](https://arxiv.org/abs/2509.04664) [find](https://scholar.google.com/scholar?q=Adam+Tauman+Kalai%2C+Ofir+Nachum%2C+Santosh+S%2E+Vempala%2C+and+Edwin+Zhang%2E+2025%2E+Why+language+models+hallucinate%2E+Preprint%2C+arXiv%3A2509%2E04664%2E+%E2%80%94+%5BarXiv%3A2509%2E04664%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2509%2E04664%29)
17. Myunsoo Kim, Hayeong Lee, Seong-Woong Shim, Junho Seo, and Byung-Jun Lee. 2025. Nbdi: A simple and effective termination condition for skill extraction from task-agnostic demonstrations. In International Conference on Machine Learning, pages 30437–30461. PMLR. [find](https://scholar.google.com/scholar?q=Myunsoo+Kim%2C+Hayeong+Lee%2C+Seong%2DWoong+Shim%2C+Junho+Seo%2C+and+Byung%2DJun+Lee%2E+2025%2E+Nbdi%3A+A+simple+and+effective+termination+condition+for+skill+extraction+from+task%2Dagnostic+demonstrations%2E+In+Internatio)
18. Heinrich Küttler, Nantas Nardelli, Alexander Miller, Roberta Raileanu, Marco Selvatici, Edward Grefenstette, and Tim Rocktäschel. 2020. The nethack learning environment. Advances in Neural Information Processing Systems, 33:7671–7684. [find](https://scholar.google.com/scholar?q=Heinrich+K%C3%BCttler%2C+Nantas+Nardelli%2C+Alexander+Miller%2C+Roberta+Raileanu%2C+Marco+Selvatici%2C+Edward+Grefenstette%2C+and+Tim+Rockt%C3%A4schel%2E+2020%2E+The+nethack+learning+environment%2E+Advances+in+Neural+Information)
19. Hayeong Lee, Jun Ho Seo, Sunguk Shin, Jinho Lee, Myunsoo Kim, Minsuk Chang, and Byung-Jun Lee. 2025. Evaluating llm planning in partially observable environments via observation representations and action sequences. In NeurIPS 2025 Workshop on Bridging Language, Agent, and World Models for Reasoning… [find](https://scholar.google.com/scholar?q=Hayeong+Lee%2C+Jun+Ho+Seo%2C+Sunguk+Shin%2C+Jinho+Lee%2C+Myunsoo+Kim%2C+Minsuk+Chang%2C+and+Byung%2DJun+Lee%2E+2025%2E+Evaluating+llm+planning+in+partially+observable+environments+via+observation+representations+and+ac)
20. 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](https://md.research.elamur.ai/free/en/papers/lewis-2020-retrieval-augmented-generation-knowledge-intensive/). Advances in neural information processing systems, 33:9459–…
21. Weiyu Ma, Qirui Mi, Yongcheng Zeng, Xue Yan, Runji Lin, Yuqiao Wu, Jun Wang, and Haifeng Zhang. 2024. Large language models play starcraft ii: Benchmarks and a chain of summarization approach. Advances in Neural Information Processing Systems, 37:133386– 133442. [find](https://scholar.google.com/scholar?q=Weiyu+Ma%2C+Qirui+Mi%2C+Yongcheng+Zeng%2C+Xue+Yan%2C+Runji+Lin%2C+Yuqiao+Wu%2C+Jun+Wang%2C+and+Haifeng+Zhang%2E+2024%2E+Large+language+models+play+starcraft+ii%3A+Benchmarks+and+a+chain+of+summarization+approach%2E+Advance)
22. The 1998 ai planning systems competition. AI Magazine, 21(2):35. [find](https://scholar.google.com/scholar?q=The+1998+ai+planning+systems+competition%2E+AI+Magazine%2C+21%282%29%3A35%2E)
23. Davide Paglieri, Bartłomiej Cupiał, Samuel Coward, Ulyana Piterbarg, Maciej Wolczyk, Akbir Khan, Eduardo Pignatelli, Łukasz Kuciński, Lerrel Pinto, Rob Fergus, and 1 others. 2024. Balrog: Benchmarking agentic llm and vlm reasoning on games. arXiv preprint arXiv:2411.13543. — [arXiv:2411.13543](https://arxiv.org/abs/2411.13543) [find](https://scholar.google.com/scholar?q=Davide+Paglieri%2C+Bart%C5%82omiej+Cupia%C5%82%2C+Samuel+Coward%2C+Ulyana+Piterbarg%2C+Maciej+Wolczyk%2C+Akbir+Khan%2C+Eduardo+Pignatelli%2C+%C5%81ukasz+Kucin%CC%81ski%2C+Lerrel+Pinto%2C+Rob+Fergus%2C+and+1+others%2E+2024%2E+Balrog%3A+Benchmarkin)
24. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. [Squad: 100,000+ questions for machine comprehension of text](https://md.research.elamur.ai/free/en/papers/rajpurkar-2016-squad-100-000-questions-machine/). arXiv preprint arXiv:1606.05250.
25. Mikayel Samvelyan, Robert Kirk, Vitaly Kurin, Jack Parker-Holder, Minqi Jiang, Eric Hambro, Fabio Petroni, Heinrich Küttler, Edward Grefenstette, and Tim Rocktäschel. 2021. Minihack the planet: A sandbox for open-ended reinforcement learning research. arXiv preprint arXiv:2109.13202. — [arXiv:2109.13202](https://arxiv.org/abs/2109.13202) [find](https://scholar.google.com/scholar?q=Mikayel+Samvelyan%2C+Robert+Kirk%2C+Vitaly+Kurin%2C+Jack+Parker%2DHolder%2C+Minqi+Jiang%2C+Eric+Hambro%2C+Fabio+Petroni%2C+Heinrich+K%C3%BCttler%2C+Edward+Grefenstette%2C+and+Tim+Rockt%C3%A4schel%2E+2021%2E+Minihack+the+planet%3A+A+sand)
26. Alberto Sanfeliu and King-Sun Fu. 2012. A distance measure between attributed relational graphs for pattern recognition. IEEE transactions on systems, man, and cybernetics, (3):353–362. [find](https://scholar.google.com/scholar?q=Alberto+Sanfeliu+and+King%2DSun+Fu%2E+2012%2E+A+distance+measure+between+attributed+relational+graphs+for+pattern+recognition%2E+IEEE+transactions+on+systems%2C+man%2C+and+cybernetics%2C+%283%29%3A353%E2%80%93362%2E)
27. Seong-Woong Shim, Myunsoo Kim, Jae Hyeon Cho, and Byung-Jun Lee. 2026. Beyond RAG vs. long-context: Learning distraction-aware retrieval for efficient knowledge grounding. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Seong%2DWoong+Shim%2C+Myunsoo+Kim%2C+Jae+Hyeon+Cho%2C+and+Byung%2DJun+Lee%2E+2026%2E+Beyond+RAG+vs%2E+long%2Dcontext%3A+Learning+distraction%2Daware+retrieval+for+efficient+knowledge+grounding%2E+In+The+Fourteenth+Internatio)
28. SungUk Shin and Youngjoon Kim. 2025. Enhancing graph of thought: Enhancing prompts with llm rationales and dynamic temperature control. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=SungUk+Shin+and+Youngjoon+Kim%2E+2025%2E+Enhancing+graph+of+thought%3A+Enhancing+prompts+with+llm+rationales+and+dynamic+temperature+control%2E+In+The+Thirteenth+International+Conference+on+Learning+Represent)
29. 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.
30. Junjie Wen, Minjie Zhu, Yichen Zhu, Zhibin Tang, Jinming Li, Zhongyi Zhou, Chengmeng Li, Xiaoyu Liu, Yaxin Peng, Chaomin Shen, and 1 others. 2024. Diffusion-vla: Generalizable and interpretable robot foundation model via self-generated reasoning. arXiv preprint arXiv:2412.03293. [find](https://scholar.google.com/scholar?q=Junjie+Wen%2C+Minjie+Zhu%2C+Yichen+Zhu%2C+Zhibin+Tang%2C+Jinming+Li%2C+Zhongyi+Zhou%2C+Chengmeng+Li%2C+Xiaoyu+Liu%2C+Yaxin+Peng%2C+Chaomin+Shen%2C+and+1+others%2E+2024%2E+Diffusion%2Dvla%3A+Generalizable+and+interpretable+robot+)
31. Jihan Yang, Shusheng Yang, Anjali W. Gupta, Rilyn Han, Li Fei-Fei, and Saining Xie. 2025. Thinking in space: How multimodal large language models see, remember, and recall spaces. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10632–10643. [find](https://scholar.google.com/scholar?q=Jihan+Yang%2C+Shusheng+Yang%2C+Anjali+W%2E+Gupta%2C+Rilyn+Han%2C+Li+Fei%2DFei%2C+and+Saining+Xie%2E+2025%2E+Thinking+in+space%3A+How+multimodal+large+language+models+see%2C+remember%2C+and+recall+spaces%2E+In+Proceedings+of+th)
32. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Tom Griffiths, Yuan Cao, and Karthik Narasimhan. 2023. Tree of thoughts: Deliberate problem solving with large language models. Advances in neural information processing systems, 36:11809–11822. [find](https://scholar.google.com/scholar?q=Shunyu+Yao%2C+Dian+Yu%2C+Jeffrey+Zhao%2C+Izhak+Shafran%2C+Tom+Griffiths%2C+Yuan+Cao%2C+and+Karthik+Narasimhan%2E+2023%2E+Tree+of+thoughts%3A+Deliberate+problem+solving+with+large+language+models%2E+Advances+in+neural+inf)
33. Wenhao Yu, Dan Iter, Shuohang Wang, Yichong Xu, Mingxuan Ju, Soumya Sanyal, Chenguang Zhu, Michael Zeng, and Meng Jiang. 2022. [Generate rather than retrieve: Large language models are strong context generators](https://md.research.elamur.ai/free/en/papers/yu-2022-generate-rather-than-retrieve-large/). arXiv preprint arXiv:2209.10063.
34. Huaixiu Steven Zheng, Swaroop Mishra, Xinyun Chen, Heng-Tze Cheng, Ed H Chi, Quoc V Le, and Denny Zhou. 2023. Take a step back: Evoking reasoning via abstraction in large language models. arXiv preprint arXiv:2310.06117. [find](https://scholar.google.com/scholar?q=Huaixiu+Steven+Zheng%2C+Swaroop+Mishra%2C+Xinyun+Chen%2C+Heng%2DTze+Cheng%2C+Ed+H+Chi%2C+Quoc+V+Le%2C+and+Denny+Zhou%2E+2023%2E+Take+a+step+back%3A+Evoking+reasoning+via+abstraction+in+large+language+models%2E+arXiv+prepri)

*Updated 1 November 2026*