---
title: "MARS-RA — paper by Wang et al. (2026) — summary and ideas"
description: "The authors also introduce MARS-Bench, a ManiSkill3-based benchmark in which agents may be added or… 70% accuracy. 67 references."
authors: ["Dawei Wang", "Di Zhao", "Xinyuan Liu", "Marci Chi Ma", "Xiaoyang Liu", "Chengming Zhou", "Gary Ushaw", "Richard Davison"]
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.1349/"
doi: "10.18653/v1/2026.acl-long.1349"
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/wang-2026-mars-ra-rank-aggregation-credit/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-mars-ra-rank-aggregation-credit/)

[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/35/) / MARS-RA: Rank Aggregation for Credit Assignment via Multimodal Comparisons in Embodied Multi-Agent Cooperation

# MARS-RA: Rank Aggregation for Credit Assignment via Multimodal Comparisons in Embodied Multi-Agent Cooperation

**Authors:** [Dawei Wang](https://md.research.elamur.ai/free/en/authors/dawei-wang/), Di Zhao, Xinyuan Liu, Marci Chi Ma, [Xiaoyang Liu](https://md.research.elamur.ai/free/en/authors/xiaoyang-liu/), Chengming Zhou, [Gary Ushaw](https://md.research.elamur.ai/free/en/authors/gary-ushaw/) and [Richard Davison](https://md.research.elamur.ai/free/en/authors/richard-davison/) · **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.1349/) · [PDF](https://aclanthology.org/2026.acl-long.1349.pdf) · [DOI 10.18653/v1/2026.acl-long.1349](https://doi.org/10.18653/v1/2026.acl-long.1349) · [OpenAlex](https://openalex.org/W7166866776)

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

## Summary
The paper addresses credit assignment in cooperative multi-agent reinforcement learning (MARL) for embodied AI. In this setting, agents see only noisy egocentric views, rewards are sparse and delayed, and agents may join or drop out mid-task. Classic methods such as VDN, QMIX and COMA struggle under these conditions.

MARS-RA treats each agent's credit as a hidden contribution score. A large multimodal model (LMM) looks at the agents' visual observations and judges which of two agents helps the team more. A rank aggregation model, by default Bradley–Terry, then turns these noisy relative judgments into continuous per-agent scores. The scores become a potential function for potential-based reward shaping, which supplies dense rewards. Because comparisons are pairwise, the method does not depend on team size.

The authors also introduce MARS-Bench, a ManiSkill3-based benchmark in which agents may be added or removed during an episode. They provide theoretical arguments for convergence and robustness, with Shapley values as an interpretive reference. MARS-RA outperforms strong baselines, and its results improve with more accurate LMMs and more comparison queries.

## Key points
- **MARS-Bench:** 2–4 decentralized agents work on three tasks (Pass Gate, Herd Sheep, Collect Ball) with partial observability, sparse rewards and agents entering or leaving.
- **Comparison accuracy:** Gemini-2.5-Pro and GPT-5.1 exceed 70% accuracy on pairwise comparisons. Qwen3-VL (2B/4B/8B) exceeds 43%. Errors cluster in cases where every agent faces a wall and no informative cues are visible.
- **Accuracy vs. query count:** more accurate LMMs and more queries both raise success rates. However, extra queries help less once the LMM is already accurate.
- **Rank aggregation choice:** Bradley–Terry slightly beats Rank Centrality.
  - Pass Gate: 0.52 vs. 0.50
  - Herd Sheep: 0.68 vs. 0.67
  - Collect Ball: 0.46 vs. 0.46
- **Real-world transfer:** policies trained in a 3D-scanned simulated room were deployed on two XLeRobot robots for Pass Gate, reaching a 64% success rate over 25 runs.

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

📚 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 · [⚙️ Engineering](https://md.research.elamur.ai/free/en/shelves/engineering/) · 133 works · [⚙️ Control and Systems Engineering](https://md.research.elamur.ai/free/en/shelves/engineering/control-and-systems-engineering/) · 45 works

🧭 Topics (1): [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 657 works

## Cite

Cite as: Dawei Wang, Di Zhao, Xinyuan Liu, Marci Chi Ma, Xiaoyang Liu, Chengming Zhou, Gary Ushaw and Richard Davison. 2026. MARS-RA: Rank Aggregation for Credit Assignment via Multimodal Comparisons in Embodied Multi-Agent Cooperation. 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{wang2026mars,
  title = {{MARS-RA: Rank Aggregation for Credit Assignment via Multimodal Comparisons in Embodied Multi-Agent Cooperation}},
  author = {Dawei Wang, Di Zhao, Xinyuan Liu, Marci Chi Ma, Xiaoyang Liu, Chengming Zhou, Gary Ushaw and Richard Davison},
  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.1349},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1349}
}
```

- [← previous Markovian Linguistic-Temporal Bridge: Unlocking the Potential of LLMs for Time Series Forecasting](https://md.research.elamur.ai/free/en/papers/sun-2026-markovian-linguistic-temporal-bridge-unlocking/)
- [next → MARS: Unleashing the Power of Speculative Decoding via Margin-Aware Verification](https://md.research.elamur.ai/free/en/papers/song-2026-mars-unleashing-power-speculative-decoding/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/35/)

## References

67 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Alireza Saleh Abadi and Leen-Kiat Soh. 2025. Challenges in credit assignment for multi-agent reinforcement learning in open agent systems. arXiv preprint arXiv:2510.27659. — [arXiv:2510.27659](https://arxiv.org/abs/2510.27659) [find](https://scholar.google.com/scholar?q=Alireza+Saleh+Abadi+and+Leen%2DKiat+Soh%2E+2025%2E+Challenges+in+credit+assignment+for+multi%2Dagent+reinforcement+learning+in+open+agent+systems%2E+arXiv+preprint+arXiv%3A2510%2E27659%2E+%E2%80%94+%5BarXiv%3A2510%2E27659%5D%28https%3A%2F)
2. Anatoly Bakushinsky and A Goncharsky. 2012. Ill-posed problems: theory and applications, volume 301. Springer Science & Business Media. [find](https://scholar.google.com/scholar?q=Anatoly+Bakushinsky+and+A+Goncharsky%2E+2012%2E+Ill%2Dposed+problems%3A+theory+and+applications%2C+volume+301%2E+Springer+Science+%26+Business+Media%2E)
3. Christopher M Bishop and Nasser M Nasrabadi. 2006. Pattern recognition and machine learning, volume 4. Springer. [find](https://scholar.google.com/scholar?q=Christopher+M+Bishop+and+Nasser+M+Nasrabadi%2E+2006%2E+Pattern+recognition+and+machine+learning%2C+volume+4%2E+Springer%2E)
4. On the utility of learning about humans for human-ai coordination. Advances in neural information processing systems, 32. [find](https://scholar.google.com/scholar?q=On+the+utility+of+learning+about+humans+for+human%2Dai+coordination%2E+Advances+in+neural+information+processing+systems%2C+32%2E)
5. Paul F Christiano, Jan Leike, Tom Brown, Miljan Martic, Shane Legg, and Dario Amodei. 2017. Deep reinforcement learning from human preferences. Advances in neural information processing systems, 30. [find](https://scholar.google.com/scholar?q=Paul+F+Christiano%2C+Jan+Leike%2C+Tom+Brown%2C+Miljan+Martic%2C+Shane+Legg%2C+and+Dario+Amodei%2E+2017%2E+Deep+reinforcement+learning+from+human+preferences%2E+Advances+in+neural+information+processing+systems%2C+30%2E)
6. Andy Clark. 1998. Being there: Putting brain, body, and world together again. MIT press. [find](https://scholar.google.com/scholar?q=Andy+Clark%2E+1998%2E+Being+there%3A+Putting+brain%2C+body%2C+and+world+together+again%2E+MIT+press%2E)
7. Jonathan Cohen, Jilles-Steeve Dibangoye, and Abdel-Illah Mouaddib. 2017. Open decentralized pomdps. In 2017 IEEE 29th International Conference on Tools with Artificial Intelligence (ICTAI), pages 977–984. IEEE. [find](https://scholar.google.com/scholar?q=Jonathan+Cohen%2C+Jilles%2DSteeve+Dibangoye%2C+and+Abdel%2DIllah+Mouaddib%2E+2017%2E+Open+decentralized+pomdps%2E+In+2017+IEEE+29th+International+Conference+on+Tools+with+Artificial+Intelligence+%28ICTAI%29%2C+pages+977%E2%80%93)
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. Douglas E Critchlow, Michael A Fligner, and Joseph S Verducci. 1991. Probability models on rankings. Journal of mathematical psychology, 35(3):294–318. [find](https://scholar.google.com/scholar?q=Douglas+E+Critchlow%2C+Michael+A+Fligner%2C+and+Joseph+S+Verducci%2E+1991%2E+Probability+models+on+rankings%2E+Journal+of+mathematical+psychology%2C+35%283%29%3A294%E2%80%93318%2E)
10. Gerard Debreu. 1960. Individual choice behavior: A theoretical analysis. [find](https://scholar.google.com/scholar?q=Gerard+Debreu%2E+1960%2E+Individual+choice+behavior%3A+A+theoretical+analysis%2E)
11. Sam Michael Devlin and Daniel Kudenko. 2012. Dynamic potential-based reward shaping. In 11th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2012), pages 433–440. IFAAMAS. [find](https://scholar.google.com/scholar?q=Sam+Michael+Devlin+and+Daniel+Kudenko%2E+2012%2E+Dynamic+potential%2Dbased+reward+shaping%2E+In+11th+International+Conference+on+Autonomous+Agents+and+Multiagent+Systems+%28AAMAS+2012%29%2C+pages+433%E2%80%93440%2E+IFAAMAS%2E)
12. Jingfeng Zhang Di Zhao, Hongsheng Hu, Philippe Fournier-Viger, Gillian Dobbie, and Yun Sing Koh. 2025. Balancing invariant and specific knowledge for domain generalization with online knowledge distillation. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligenc… [find](https://scholar.google.com/scholar?q=Jingfeng+Zhang+Di+Zhao%2C+Hongsheng+Hu%2C+Philippe+Fournier%2DViger%2C+Gillian+Dobbie%2C+and+Yun+Sing+Koh%2E+2025%2E+Balancing+invariant+and+specific+knowledge+for+domain+generalization+with+online+knowledge+distil)
13. Shuai Dong, Siyuan Wang, Xingyu Liu, Chenglin Li, Haowen Hou, and Zhongyu Wei. 2025. [Interleaved latent visual reasoning with selective perceptual modeling](https://md.research.elamur.ai/free/en/papers/dong-2026-interleaved-latent-visual-reasoning-selective/). arXiv preprint arXiv:2512.05665.
14. Tongtong Feng, Xin Wang, Yu-Gang Jiang, and Wenwu Zhu. 2025. Embodied ai: From llms to world models. arXiv preprint arXiv:2509.20021. — [arXiv:2509.20021](https://arxiv.org/abs/2509.20021) [find](https://scholar.google.com/scholar?q=Tongtong+Feng%2C+Xin+Wang%2C+Yu%2DGang+Jiang%2C+and+Wenwu+Zhu%2E+2025%2E+Embodied+ai%3A+From+llms+to+world+models%2E+arXiv+preprint+arXiv%3A2509%2E20021%2E+%E2%80%94+%5BarXiv%3A2509%2E20021%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2509%2E20021%29)
15. Ralph Allan Bradley and Milton E Terry. 1952. Rank analysis of incomplete block designs: I. the method of paired comparisons. Biometrika, 39(3/4):324– 345. [find](https://scholar.google.com/scholar?q=Ralph+Allan+Bradley+and+Milton+E+Terry%2E+1952%2E+Rank+analysis+of+incomplete+block+designs%3A+I%2E+the+method+of+paired+comparisons%2E+Biometrika%2C+39%283%2F4%29%3A324%E2%80%93+345%2E)
16. Jakob Foerster, Gregory Farquhar, Triantafyllos Afouras, Nantas Nardelli, and Shimon Whiteson. 2017. Counterfactual multi-agent policy gradients (2017). arXiv preprint arXiv:1705.08926. — [arXiv:1705.08926](https://arxiv.org/abs/1705.08926) [find](https://scholar.google.com/scholar?q=Jakob+Foerster%2C+Gregory+Farquhar%2C+Triantafyllos+Afouras%2C+Nantas+Nardelli%2C+and+Shimon+Whiteson%2E+2017%2E+Counterfactual+multi%2Dagent+policy+gradients+%282017%29%2E+arXiv+preprint+arXiv%3A1705%2E08926%2E+%E2%80%94+%5BarXiv%3A1705%2E)
17. Rodney A Brooks. 1991. New approaches to robotics. Science, 253(5025):1227–1232. [find](https://scholar.google.com/scholar?q=Rodney+A+Brooks%2E+1991%2E+New+approaches+to+robotics%2E+Science%2C+253%285025%29%3A1227%E2%80%931232%2E)
18. Anna Harutyunyan, Sam Devlin, Peter Vrancx, and Ann Nowé. 2015. Expressing arbitrary reward functions as potential-based advice. In Proceedings of the AAAI conference on artificial intelligence, volume 29. [find](https://scholar.google.com/scholar?q=Anna+Harutyunyan%2C+Sam+Devlin%2C+Peter+Vrancx%2C+and+Ann+Now%C3%A9%2E+2015%2E+Expressing+arbitrary+reward+functions+as+potential%2Dbased+advice%2E+In+Proceedings+of+the+AAAI+conference+on+artificial+intelligence%2C+volum)
19. Micah Carroll, Rohin Shah, Mark K Ho, Tom Griffiths, Sanjit Seshia, Pieter Abbeel, and Anca Dragan. 2019. [find](https://scholar.google.com/scholar?q=Micah+Carroll%2C+Rohin+Shah%2C+Mark+K+Ho%2C+Tom+Griffiths%2C+Sanjit+Seshia%2C+Pieter+Abbeel%2C+and+Anca+Dragan%2E+2019%2E)
20. Ralf Herbrich, Tom Minka, and Thore Graepel. 2006. Trueskill™: a bayesian skill rating system. Advances in neural information processing systems, 19. [find](https://scholar.google.com/scholar?q=Ralf+Herbrich%2C+Tom+Minka%2C+and+Thore+Graepel%2E+2006%2E+Trueskill%E2%84%A2%3A+a+bayesian+skill+rating+system%2E+Advances+in+neural+information+processing+systems%2C+19%2E)
21. Shaohan Huang, Li Dong, Wenhui Wang, Yaru Hao, Saksham Singhal, Shuming Ma, Tengchao Lv, Lei Cui, Owais Khan Mohammed, Barun Patra, and 1 others. 2023. [Language is not all you need: Aligning perception with language models](https://md.research.elamur.ai/free/en/papers/huang-2023-language-all-need-aligning-perception/). Advances in Neural Information Processing Systems, 36:72096– 72109.
22. Raivo Kolde, Sven Laur, Priit Adler, and Jaak Vilo. 2012. Robust rank aggregation for gene list integration and meta-analysis. Bioinformatics, 28(4):573–580. [find](https://scholar.google.com/scholar?q=Raivo+Kolde%2C+Sven+Laur%2C+Priit+Adler%2C+and+Jaak+Vilo%2E+2012%2E+Robust+rank+aggregation+for+gene+list+integration+and+meta%2Danalysis%2E+Bioinformatics%2C+28%284%29%3A573%E2%80%93580%2E)
23. Chenglin Li, Qianglong Chen, Feng Han, Yikun Wang, Xingxi Yin, Yan Gong, Ruilin Li, Yin Zhang, and Jiaqi Wang. 2026. Videothinker: Building agentic videollms with llm-guided tool reasoning. arXiv preprint arXiv:2601.15724. — [arXiv:2601.15724](https://arxiv.org/abs/2601.15724) [find](https://scholar.google.com/scholar?q=Chenglin+Li%2C+Qianglong+Chen%2C+Feng+Han%2C+Yikun+Wang%2C+Xingxi+Yin%2C+Yan+Gong%2C+Ruilin+Li%2C+Yin+Zhang%2C+and+Jiaqi+Wang%2E+2026%2E+Videothinker%3A+Building+agentic+videollms+with+llm%2Dguided+tool+reasoning%2E+arXiv+prep)
24. Chenglin Li, Feng Han, Yikun Wang, Ruilin Li, Shuai Dong, Haowen Hou, Haitao Li, Qianglong Chen, Feng Tao, Jingqi Tong, and 1 others. 2025a. Video-pro: Adaptive program reasoning for long video understanding. arXiv preprint arXiv:2509.17743. — [arXiv:2509.17743](https://arxiv.org/abs/2509.17743) [find](https://scholar.google.com/scholar?q=Chenglin+Li%2C+Feng+Han%2C+Yikun+Wang%2C+Ruilin+Li%2C+Shuai+Dong%2C+Haowen+Hou%2C+Haitao+Li%2C+Qianglong+Chen%2C+Feng+Tao%2C+Jingqi+Tong%2C+and+1+others%2E+2025a%2E+Video%2Dpro%3A+Adaptive+program+reasoning+for+long+video+unders)
25. Wenhao Li, Dan Qiao, Baoxiang Wang, Xiangfeng Wang, Wei Yin, Hao Shen, Bo Jin, and Hongyuan Zha. 2025b. Multi-agent credit assignment with pre-trained language models. In International Conference on Artificial Intelligence and Statistics, pages 1945–1953. PMLR. [find](https://scholar.google.com/scholar?q=Wenhao+Li%2C+Dan+Qiao%2C+Baoxiang+Wang%2C+Xiangfeng+Wang%2C+Wei+Yin%2C+Hao+Shen%2C+Bo+Jin%2C+and+Hongyuan+Zha%2E+2025b%2E+Multi%2Dagent+credit+assignment+with+pre%2Dtrained+language+models%2E+In+International+Conference+on+A)
26. Yunxin Li, Baotian Hu, Xinyu Chen, Lin Ma, Yong Xu, and Min Zhang. 2024. Lmeye: An interactive perception network for large language models. IEEE Transactions on Multimedia. [find](https://scholar.google.com/scholar?q=Yunxin+Li%2C+Baotian+Hu%2C+Xinyu+Chen%2C+Lin+Ma%2C+Yong+Xu%2C+and+Min+Zhang%2E+2024%2E+Lmeye%3A+An+interactive+perception+network+for+large+language+models%2E+IEEE+Transactions+on+Multimedia%2E)
27. Muhan Lin, Shuyang Shi, Yue Guo, Vaishnav Tadiparthi, Behdad Chalaki, Ehsan Moradi Pari, Simon Stepputtis, Woojun Kim, Joseph Campbell, and Katia Sycara. 2025. Speaking the language of teamwork: Llm-guided credit assignment in multi-agent reinforcement learning. arXiv preprint arXiv:2502.03723. — [arXiv:2502.03723](https://arxiv.org/abs/2502.03723) [find](https://scholar.google.com/scholar?q=Muhan+Lin%2C+Shuyang+Shi%2C+Yue+Guo%2C+Vaishnav+Tadiparthi%2C+Behdad+Chalaki%2C+Ehsan+Moradi+Pari%2C+Simon+Stepputtis%2C+Woojun+Kim%2C+Joseph+Campbell%2C+and+Katia+Sycara%2E+2025%2E+Speaking+the+language+of+teamwork%3A+Llm%2Dg)
28. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023a. [Visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/). Advances in neural information processing systems, 36:34892– 34916.
29. Wenzhang Liu, Wenzhe Cai, Kun Jiang, Guangran Cheng, Yuanda Wang, Jiawei Wang, Jingyu Cao, Lele Xu, Chaoxu Mu, and Changyin Sun. 2023b. Xuance: A comprehensive and unified deep reinforcement learning library. arXiv preprint arXiv:2312.16248. [find](https://scholar.google.com/scholar?q=Wenzhang+Liu%2C+Wenzhe+Cai%2C+Kun+Jiang%2C+Guangran+Cheng%2C+Yuanda+Wang%2C+Jiawei+Wang%2C+Jingyu+Cao%2C+Lele+Xu%2C+Chaoxu+Mu%2C+and+Changyin+Sun%2E+2023b%2E+Xuance%3A+A+comprehensive+and+unified+deep+reinforcement+learning+)
30. Hao Ma, Shijie Wang, Zhiqiang Pu, Siyao Zhao, and Xiaolin Ai. 2025. Vision-based generic potential function for policy alignment in multi-agent reinforcement learning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 19287–19295. [find](https://scholar.google.com/scholar?q=Hao+Ma%2C+Shijie+Wang%2C+Zhiqiang+Pu%2C+Siyao+Zhao%2C+and+Xiaolin+Ai%2E+2025%2E+Vision%2Dbased+generic+potential+function+for+policy+alignment+in+multi%2Dagent+reinforcement+learning%2E+In+Proceedings+of+the+AAAI+Confe)
31. Ke Ma, Qianqian Xu, Jinshan Zeng, Guorong Li, Xiaochun Cao, and Qingming Huang. 2022. A tale of hodgerank and spectral method: Target attack against rank aggregation is the fixed point of adversarial game. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(4):4090–4108. [find](https://scholar.google.com/scholar?q=Ke+Ma%2C+Qianqian+Xu%2C+Jinshan+Zeng%2C+Guorong+Li%2C+Xiaochun+Cao%2C+and+Qingming+Huang%2E+2022%2E+A+tale+of+hodgerank+and+spectral+method%3A+Target+attack+against+rank+aggregation+is+the+fixed+point+of+adversarial+)
32. Marvin Minsky. 2007. Steps toward artificial intelligence. Proceedings of the IRE, 49(1):8–30. [find](https://scholar.google.com/scholar?q=Marvin+Minsky%2E+2007%2E+Steps+toward+artificial+intelligence%2E+Proceedings+of+the+IRE%2C+49%281%29%3A8%E2%80%9330%2E)
33. Kartik Nagpal, Dayi Dong, Jean-Baptiste Bouvier, and Negar Mehr. 2025. Leveraging large language models for effective and explainable multi-agent credit assignment. arXiv preprint arXiv:2502.16863. — [arXiv:2502.16863](https://arxiv.org/abs/2502.16863) [find](https://scholar.google.com/scholar?q=Kartik+Nagpal%2C+Dayi+Dong%2C+Jean%2DBaptiste+Bouvier%2C+and+Negar+Mehr%2E+2025%2E+Leveraging+large+language+models+for+effective+and+explainable+multi%2Dagent+credit+assignment%2E+arXiv+preprint+arXiv%3A2502%2E16863%2E+%E2%80%94+)
34. Sahand Negahban, Sewoong Oh, and Devavrat Shah. 2012a. Iterative ranking from pair-wise comparisons. Advances in neural information processing systems, 25. [find](https://scholar.google.com/scholar?q=Sahand+Negahban%2C+Sewoong+Oh%2C+and+Devavrat+Shah%2E+2012a%2E+Iterative+ranking+from+pair%2Dwise+comparisons%2E+Advances+in+neural+information+processing+systems%2C+25%2E)
35. Sahand N Negahban, Pradeep Ravikumar, Martin J Wainwright, and Bin Yu. 2012b. A unified framework for high-dimensional analysis of m-estimators with decomposable regularizers. Statistical Science, 27(4):538–557. [find](https://scholar.google.com/scholar?q=Sahand+N+Negahban%2C+Pradeep+Ravikumar%2C+Martin+J+Wainwright%2C+and+Bin+Yu%2E+2012b%2E+A+unified+framework+for+high%2Ddimensional+analysis+of+m%2Destimators+with+decomposable+regularizers%2E+Statistical+Science%2C+27%28)
36. Andrew Y Ng, Daishi Harada, and Stuart Russell. 1999. Policy invariance under reward transformations: Theory and application to reward shaping. In Icml, volume 99, pages 278–287. Citeseer. [find](https://scholar.google.com/scholar?q=Andrew+Y+Ng%2C+Daishi+Harada%2C+and+Stuart+Russell%2E+1999%2E+Policy+invariance+under+reward+transformations%3A+Theory+and+application+to+reward+shaping%2E+In+Icml%2C+volume+99%2C+pages+278%E2%80%93287%2E+Citeseer%2E)
37. OpenAI. 2025. Gpt-5 system card. [https://cdn.openai.com/gpt-5-system-card.pdf](https://cdn.openai.com/gpt-5-system-card.pdf). [↗](https://cdn.openai.com/gpt-5-system-card.pdf)
38. Georgios Papoudakis, Filippos Christianos, Lukas Schäfer, and Stefano V Albrecht. 2020. Benchmarking multi-agent deep reinforcement learning algorithms in cooperative tasks. arXiv preprint arXiv:2006.07869. — [arXiv:2006.07869](https://arxiv.org/abs/2006.07869) [find](https://scholar.google.com/scholar?q=Georgios+Papoudakis%2C+Filippos+Christianos%2C+Lukas+Sch%C3%A4fer%2C+and+Stefano+V+Albrecht%2E+2020%2E+Benchmarking+multi%2Dagent+deep+reinforcement+learning+algorithms+in+cooperative+tasks%2E+arXiv+preprint+arXiv%3A2006%2E)
39. Dipam Patel, Phu Pham, Kshitij Tiwari, and Aniket Bera. 2023. Dream: Decentralized reinforcement learning for exploration and efficient energy management in multi-robot systems. arXiv preprint arXiv:2309.17433. — [arXiv:2309.17433](https://arxiv.org/abs/2309.17433) [find](https://scholar.google.com/scholar?q=Dipam+Patel%2C+Phu+Pham%2C+Kshitij+Tiwari%2C+and+Aniket+Bera%2E+2023%2E+Dream%3A+Decentralized+reinforcement+learning+for+exploration+and+efficient+energy+management+in+multi%2Drobot+systems%2E+arXiv+preprint+arXiv%3A2)
40. Tabish Rashid, Mikayel Samvelyan, Christian Schroeder De Witt, Gregory Farquhar, Jakob Foerster, and Shimon Whiteson. 2020. Monotonic value function factorisation for deep multi-agent reinforcement learning. Journal of Machine Learning Research, 21(178):1–51. [find](https://scholar.google.com/scholar?q=Tabish+Rashid%2C+Mikayel+Samvelyan%2C+Christian+Schroeder+De+Witt%2C+Gregory+Farquhar%2C+Jakob+Foerster%2C+and+Shimon+Whiteson%2E+2020%2E+Monotonic+value+function+factorisation+for+deep+multi%2Dagent+reinforcement+le)
41. Juan Rocamonde, Victoriano Montesinos, Elvis Nava, Ethan Perez, and David Lindner. 2023. Vision-language models are zero-shot reward models for reinforcement learning. arXiv preprint arXiv:2310.12921. [find](https://scholar.google.com/scholar?q=Juan+Rocamonde%2C+Victoriano+Montesinos%2C+Elvis+Nava%2C+Ethan+Perez%2C+and+David+Lindner%2E+2023%2E+Vision%2Dlanguage+models+are+zero%2Dshot+reward+models+for+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2310%2E12921%2E)
42. Álvaro Serra-Gómez, Hai Zhu, Bruno Brito, Wendelin Böhmer, and Javier Alonso-Mora. 2023. Learning scalable and efficient communication policies for multi-robot collision avoidance. Autonomous Robots, 47(8):1275–1297. [find](https://scholar.google.com/scholar?q=%C3%81lvaro+Serra%2DG%C3%B3mez%2C+Hai+Zhu%2C+Bruno+Brito%2C+Wendelin+B%C3%B6hmer%2C+and+Javier+Alonso%2DMora%2E+2023%2E+Learning+scalable+and+efficient+communication+policies+for+multi%2Drobot+collision+avoidance%2E+Autonomous+Robots%2C+)
43. Nihar B Shah, Sivaraman Balakrishnan, Joseph Bradley, Abhay Parekh, Kannan Ramchandran, and Martin J Wainwright. 2016. Estimation from pairwise comparisons: Sharp minimax bounds with topology dependence. Journal of Machine Learning Research, 17(58):1–47. [find](https://scholar.google.com/scholar?q=Nihar+B+Shah%2C+Sivaraman+Balakrishnan%2C+Joseph+Bradley%2C+Abhay+Parekh%2C+Kannan+Ramchandran%2C+and+Martin+J+Wainwright%2E+2016%2E+Estimation+from+pairwise+comparisons%3A+Sharp+minimax+bounds+with+topology+dependen)
44. Lloyd S Shapley and 1 others. 1953. A value for n-person games. [find](https://scholar.google.com/scholar?q=Lloyd+S+Shapley+and+1+others%2E+1953%2E+A+value+for+n%2Dperson+games%2E)
45. Lin Shi, Chiyu Ma, Wenhua Liang, Xingjian Diao, Weicheng Ma, and Soroush Vosoughi. 2024. [Judging the judges: A systematic study of position bias in llm-as-a-judge](https://md.research.elamur.ai/free/en/papers/shi-2025-judging-judges-systematic-study-position/). arXiv preprint arXiv:2406.07791.
46. Peter Sunehag, Guy Lever, Audrunas Gruslys, Wojciech Marian Czarnecki, Vinicius Zambaldi, Max Jaderberg, Marc Lanctot, Nicolas Sonnerat, Joel Z Leibo, Karl Tuyls, and 1 others. 2017. Value-decomposition networks for cooperative multi-agent learning. arXiv preprint arXiv:1706.05296. — [arXiv:1706.05296](https://arxiv.org/abs/1706.05296) [find](https://scholar.google.com/scholar?q=Peter+Sunehag%2C+Guy+Lever%2C+Audrunas+Gruslys%2C+Wojciech+Marian+Czarnecki%2C+Vinicius+Zambaldi%2C+Max+Jaderberg%2C+Marc+Lanctot%2C+Nicolas+Sonnerat%2C+Joel+Z+Leibo%2C+Karl+Tuyls%2C+and+1+others%2E+2017%2E+Value%2Ddecompositi)
47. Xuting Tang, Jia Xu, and Shusen Wang. 2023. Roma: Resilient multi-agent reinforcement learning with dynamic participating agents. In 2023 IEEE 12th International Conference on Cloud Networking (Cloud-Net), pages 247–255. IEEE. [find](https://scholar.google.com/scholar?q=Xuting+Tang%2C+Jia+Xu%2C+and+Shusen+Wang%2E+2023%2E+Roma%3A+Resilient+multi%2Dagent+reinforcement+learning+with+dynamic+participating+agents%2E+In+2023+IEEE+12th+International+Conference+on+Cloud+Networking+%28Cloud%2D)
48. Stone Tao, Fanbo Xiang, Arth Shukla, Yuzhe Qin, Xander Hinrichsen, Xiaodi Yuan, Chen Bao, Xinsong Lin, Yulin Liu, Tse-kai Chan, and 1 others. 2024. Maniskill3: Gpu parallelized robotics simulation and rendering for generalizable embodied ai. arXiv preprint arXiv:2410.00425. [find](https://scholar.google.com/scholar?q=Stone+Tao%2C+Fanbo+Xiang%2C+Arth+Shukla%2C+Yuzhe+Qin%2C+Xander+Hinrichsen%2C+Xiaodi+Yuan%2C+Chen+Bao%2C+Xinsong+Lin%2C+Yulin+Liu%2C+Tse%2Dkai+Chan%2C+and+1+others%2E+2024%2E+Maniskill3%3A+Gpu+parallelized+robotics+simulation+and)
49. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, and 1 others. 2023. [Gemini: a family of highly capable multimodal models](https://md.research.elamur.ai/free/en/papers/gemini-team-2023-gemini-family-highly-capable-multimodal/). arXiv preprint arXiv:2312.11805.
50. J Terry, Benjamin Black, Nathaniel Grammel, Mario Jayakumar, Ananth Hari, Ryan Sullivan, Luis S Santos, Clemens Dieffendahl, Caroline Horsch, Rodrigo Perez-Vicente, and 1 others. 2021. Pettingzoo: Gym for multi-agent reinforcement learning. Advances in Neural Information Processing Systems, 34:15032… [find](https://scholar.google.com/scholar?q=J+Terry%2C+Benjamin+Black%2C+Nathaniel+Grammel%2C+Mario+Jayakumar%2C+Ananth+Hari%2C+Ryan+Sullivan%2C+Luis+S+Santos%2C+Clemens+Dieffendahl%2C+Caroline+Horsch%2C+Rodrigo+Perez%2DVicente%2C+and+1+others%2E+2021%2E+Pettingzoo%3A+Gym)
51. Xinyu Tian, Shu Zou, Zhaoyuan Yang, and Jing Zhang. 2025. Identifying and mitigating position bias of multi-image vision-language models. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 10599–10609. [find](https://scholar.google.com/scholar?q=Xinyu+Tian%2C+Shu+Zou%2C+Zhaoyuan+Yang%2C+and+Jing+Zhang%2E+2025%2E+Identifying+and+mitigating+position+bias+of+multi%2Dimage+vision%2Dlanguage+models%2E+In+Proceedings+of+the+Computer+Vision+and+Pattern+Recognition+)
52. Alan M Turing. 2021. Computing machinery and intelligence (1950). Mind, 59(236):33–60. [find](https://scholar.google.com/scholar?q=Alan+M+Turing%2E+2021%2E+Computing+machinery+and+intelligence+%281950%29%2E+Mind%2C+59%28236%29%3A33%E2%80%9360%2E)
53. Dawei Wang, Chengming Zhou, Di Zhao, Xinyuan Liu, Marci Chi Ma, Gary Ushaw, and Richard Davison. 2026. Towermind: A tower defence game learning environment and benchmark for llm as agents. arXiv preprint arXiv:2601.05899. — [arXiv:2601.05899](https://arxiv.org/abs/2601.05899) [find](https://scholar.google.com/scholar?q=Dawei+Wang%2C+Chengming+Zhou%2C+Di+Zhao%2C+Xinyuan+Liu%2C+Marci+Chi+Ma%2C+Gary+Ushaw%2C+and+Richard+Davison%2E+2026%2E+Towermind%3A+A+tower+defence+game+learning+environment+and+benchmark+for+llm+as+agents%2E+arXiv+prepr)
54. Gaotian Wang and Zhuoyi Lu. 2025. Xlerobot: A practical low-cost household dual-arm mobile robot design for general manipulation. [https://github.com/](https://github.com/) Vector-Wangel/XLeRobot. [↗](https://github.com/)
55. Jianhong Wang, Yuan Zhang, Yunjie Gu, and Tae-Kyun Kim. 2022. Shaq: Incorporating shapley value theory into multi-agent q-learning. Advances in Neural Information Processing Systems, 35:5941–5954. [find](https://scholar.google.com/scholar?q=Jianhong+Wang%2C+Yuan+Zhang%2C+Yunjie+Gu%2C+and+Tae%2DKyun+Kim%2E+2022%2E+Shaq%3A+Incorporating+shapley+value+theory+into+multi%2Dagent+q%2Dlearning%2E+Advances+in+Neural+Information+Processing+Systems%2C+35%3A5941%E2%80%935954%2E)
56. Jianhong Wang, Yuan Zhang, Tae-Kyun Kim, and Yunjie Gu. 2020. Shapley q-value: A local reward approach to solve global reward games. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 7285–7292. [find](https://scholar.google.com/scholar?q=Jianhong+Wang%2C+Yuan+Zhang%2C+Tae%2DKyun+Kim%2C+and+Yunjie+Gu%2E+2020%2E+Shapley+q%2Dvalue%3A+A+local+reward+approach+to+solve+global+reward+games%2E+In+Proceedings+of+the+AAAI+conference+on+artificial+intelligence%2C+v)
57. Yuan Wei, Xiaohan Shan, Ran Miao, and Jianmin Li. 2025. Lero: Llm-driven evolutionary framework with hybrid rewards and enhanced observation for multi-agent reinforcement learning. In International Conference on Intelligent Computing, pages 15–26. Springer. [find](https://scholar.google.com/scholar?q=Yuan+Wei%2C+Xiaohan+Shan%2C+Ran+Miao%2C+and+Jianmin+Li%2E+2025%2E+Lero%3A+Llm%2Ddriven+evolutionary+framework+with+hybrid+rewards+and+enhanced+observation+for+multi%2Dagent+reinforcement+learning%2E+In+International+Co)
58. Daan Wierstra, Tom Schaul, Jan Peters, and Juergen Schmidhuber. 2008. Episodic reinforcement learning by logistic reward-weighted regression. In International Conference on Artificial Neural Networks, pages 407–416. Springer. [find](https://scholar.google.com/scholar?q=Daan+Wierstra%2C+Tom+Schaul%2C+Jan+Peters%2C+and+Juergen+Schmidhuber%2E+2008%2E+Episodic+reinforcement+learning+by+logistic+reward%2Dweighted+regression%2E+In+International+Conference+on+Artificial+Neural+Networks%2C)
59. Annie Wong, Thomas Bäck, Anna V Kononova, and Aske Plaat. 2023. Deep multiagent reinforcement learning: Challenges and directions. Artificial Intelligence Review, 56(6):5023–5056. [find](https://scholar.google.com/scholar?q=Annie+Wong%2C+Thomas+B%C3%A4ck%2C+Anna+V+Kononova%2C+and+Aske+Plaat%2E+2023%2E+Deep+multiagent+reinforcement+learning%3A+Challenges+and+directions%2E+Artificial+Intelligence+Review%2C+56%286%29%3A5023%E2%80%935056%2E)
60. Sarah A Wu, Rose E Wang, James A Evans, Joshua B Tenenbaum, David C Parkes, and Max Kleiman-Weiner. 2021. Too many cooks: Bayesian inference for coordinating multi-agent collaboration. Topics in Cognitive Science, 13(2):414–432. [find](https://scholar.google.com/scholar?q=Sarah+A+Wu%2C+Rose+E+Wang%2C+James+A+Evans%2C+Joshua+B+Tenenbaum%2C+David+C+Parkes%2C+and+Max+Kleiman%2DWeiner%2E+2021%2E+Too+many+cooks%3A+Bayesian+inference+for+coordinating+multi%2Dagent+collaboration%2E+Topics+in+Cogni)
61. Ziyan Xiong, Bo Chen, Shiyu Huang, Wei-Wei Tu, Zhaofeng He, and Yang Gao. 2024. Mqe: Unleashing the power of interaction with multi-agent quadruped environment. In 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 5918–5924. IEEE. [find](https://scholar.google.com/scholar?q=Ziyan+Xiong%2C+Bo+Chen%2C+Shiyu+Huang%2C+Wei%2DWei+Tu%2C+Zhaofeng+He%2C+and+Yang+Gao%2E+2024%2E+Mqe%3A+Unleashing+the+power+of+interaction+with+multi%2Dagent+quadruped+environment%2E+In+2024+IEEE%2FRSJ+International+Conferen)
62. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
63. Chao Yu, Akash Velu, Eugene Vinitsky, Jiaxuan Gao, Yu Wang, Alexandre Bayen, and Yi Wu. 2022. The surprising effectiveness of ppo in cooperative multi-agent games. Advances in neural information processing systems, 35:24611–24624. [find](https://scholar.google.com/scholar?q=Chao+Yu%2C+Akash+Velu%2C+Eugene+Vinitsky%2C+Jiaxuan+Gao%2C+Yu+Wang%2C+Alexandre+Bayen%2C+and+Yi+Wu%2E+2022%2E+The+surprising+effectiveness+of+ppo+in+cooperative+multi%2Dagent+games%2E+Advances+in+neural+information+proce)
64. Hongxin Zhang, Zeyuan Wang, Qiushi Lyu, Zheyuan Zhang, Sunli Chen, Tianmin Shu, Behzad Dar-iush, Kwonjoon Lee, Yilun Du, and Chuang Gan. 2024. Combo: compositional world models for embodied multi-agent cooperation. arXiv preprint arXiv:2404.10775. — [arXiv:2404.10775](https://arxiv.org/abs/2404.10775) [find](https://scholar.google.com/scholar?q=Hongxin+Zhang%2C+Zeyuan+Wang%2C+Qiushi+Lyu%2C+Zheyuan+Zhang%2C+Sunli+Chen%2C+Tianmin+Shu%2C+Behzad+Dar%2Diush%2C+Kwonjoon+Lee%2C+Yilun+Du%2C+and+Chuang+Gan%2E+2024%2E+Combo%3A+compositional+world+models+for+embodied+multi%2Dagen)
65. Kaiqing Zhang, Zhuoran Yang, and Tamer Başar. 2021. Multi-agent reinforcement learning: A selective overview of theories and algorithms. Handbook of reinforcement learning and control, pages 321–384. [find](https://scholar.google.com/scholar?q=Kaiqing+Zhang%2C+Zhuoran+Yang%2C+and+Tamer+Bas%CC%A7ar%2E+2021%2E+Multi%2Dagent+reinforcement+learning%3A+A+selective+overview+of+theories+and+algorithms%2E+Handbook+of+reinforcement+learning+and+control%2C+pages+321%E2%80%93384%2E)
66. Di Zhao, Yun Sing Koh, Gillian Dobbie, Hongsheng Hu, and Philippe Fournier-Viger. 2024. Symmetric self-paced learning for domain generalization. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 16961–16969. [find](https://scholar.google.com/scholar?q=Di+Zhao%2C+Yun+Sing+Koh%2C+Gillian+Dobbie%2C+Hongsheng+Hu%2C+and+Philippe+Fournier%2DViger%2E+2024%2E+Symmetric+self%2Dpaced+learning+for+domain+generalization%2E+In+Proceedings+of+the+AAAI+Conference+on+Artificial+Int)
67. Di Zhao, Jingfeng Zhang, Hongsheng Hu, Philippe Fournier-Viger, Gillian Dobbie, and Yun Sing Koh. 2026. Unlearning during training: Domain-specific gradient ascent for domain generalization. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Di+Zhao%2C+Jingfeng+Zhang%2C+Hongsheng+Hu%2C+Philippe+Fournier%2DViger%2C+Gillian+Dobbie%2C+and+Yun+Sing+Koh%2E+2026%2E+Unlearning+during+training%3A+Domain%2Dspecific+gradient+ascent+for+domain+generalization%2E+In+The+Fo)

*Updated 1 November 2026*