Elamur Research⌕
☰
home / papers / 2026 / MARCH: Evaluating the Intersection of Ambiguity Interpretation and Multi-hop Inference

MARCH: Evaluating the Intersection of Ambiguity Interpretation and Multi-hop Inference

Paper · 2026 · Jeonghyun Park, Ingeol Baek, Seunghyun Yoon et al.

Summary

Benchmarks for question answering usually test ambiguity or multi-hop reasoning, but rarely both together. This paper studies the combined case. When a question needs several reasoning steps, an unclear term at an early hop can hide a second ambiguity later in the chain. Models that pick one reading too soon then lose valid answers.

The authors build MARCH from ambiguous questions in MuSiQue using a four-stage pipeline:

  1. Multi-LLM ambiguity detection with clarifying-question generation
  2. Evidence document collection and reranking
  3. Short answers for each interpretation, plus a synthesized long answer
  4. Filtering by humans and LLMs

They classify the ambiguity as semantic, syntactic or constraint-based. Strong current models perform reasonably on ASQA or MuSiQue alone but degrade sharply on MARCH, often giving one-sided answers.

As a remedy they propose CLARION, an agentic method in two stages. It first plans the possible interpretations, then retrieves and reasons over evidence separately for each one. It clearly beats existing baselines. The work shows that this kind of path-dependent ambiguity is common in real user queries and remains largely unsolved.

Key points

  • MARCH has 2,209 ambiguous multi-hop questions taken from MuSiQue's validation set and part of its training set. Each comes with clarified questions, per-interpretation short answers, supporting passages and a long answer.
  • The taxonomy has three types, each paired with a model action: semantic (interpret), syntactic (resolve) and constraint (generalize).
  • Five annotators reviewed a stratified sample. Over 90% of long answers covered every interpretation, and Fleiss' κ reached 0.95.
  • In an analysis of lmsys-chat-1m queries, 48.4% were ambiguous, 17.7% were multi-hop and 13.3% were both.
  • CLARION uses a Planning Agent to lay out interpretations before evidence-driven acting, which prevents early pruning of branches. The cost is extra system complexity.

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

🏢Adobe SystemsSan Jose · United States35 works · 3,711 citations🎓Chung-Ang UniversitySeoul · South Korea3 works · 12 citations

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 worksInformation Systems · 1,437 works

Cite

Cite as Jeonghyun Park, Ingeol Baek, Seunghyun Yoon, Haeun Jang, Aparna Garimella, Akriti Jain, Nedim Lipka and Hwanhee Lee. 2026. MARCH: Evaluating the Intersection of Ambiguity Interpretation and Multi-hop Inference. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.1352

@inproceedings{park2026march,
  title = {{MARCH: Evaluating the Intersection of Ambiguity Interpretation and Multi-hop Inference}},
  author = {Jeonghyun Park, Ingeol Baek, Seunghyun Yoon, Haeun Jang, Aparna Garimella, Akriti Jain, Nedim Lipka and Hwanhee Lee},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1352},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1352}
}

References

30 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Meta AI. 2025. Llama 4 maverick: A 17b-128e multimodal language model. https://ai.meta.com/blog/llama-4-multimodal-intelligence/. Available from Meta AI Blog, April 5, 2025. ↗
  3. Anthropic. 2025. Claude sonnet 4. https://www.anthropic.com/claude/sonnet. Available on Claude Developer Platform / Anthropic site. ↗
  4. 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. arX…
  5. Jie He, Nan Hu, Wanqiu Long, Jiaoyan Chen, and Jeff Z Pan. 2024. Mintqa: A multi-hop question answering benchmark for evaluating llms on new and tail knowledge. arXiv preprint arXiv:2412.17032. — [arXiv:2412.17032](https://arxiv.org/abs/2412.17032) find
  6. Xanh Ho, Anh-Khoa Duong Nguyen, Saku Sugawara, and Akiko Aizawa. 2020a. Constructing a multi-hop QA dataset for comprehensive evaluation of reasoning steps. In Proceedings of the 28th International Conference on Computational Linguistics, pages 6609–6625, Barcelona, Spain (Online). International Com…
  7. Xanh Ho, Anh-Khoa Duong Nguyen, Saku Sugawara, and Akiko Aizawa. 2020b. Constructing a multi-hop qa dataset for comprehensive evaluation of reasoning steps. In Proceedings of the 28th International Conference on Computational Linguistics, pages 6609– 6625.
  8. Yeonjun In, Sungchul Kim, Ryan A. Rossi, Mehrab Tanjim, Tong Yu, Ritwik Sinha, and Chanyoung Park. 2025. Diversify-verify-adapt: Efficient and robust retrieval-augmented ambiguous question answering. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for… find
  9. Gangwoo Kim, Sungdong Kim, Byeongguk Jeon, Joonsuk Park, and Jaewoo Kang. 2023. Tree of clarifications: Answering ambiguous questions with retrieval-augmented large language models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 996–1009, Singapore.…
  10. Yoonsang Lee, Xi Ye, and Eunsol Choi. 2024. Ambigdocs: Reasoning across documents on different entities under the same name. In First Conference on Language Modeling. find
  11. 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. Advances in neural information processing systems, 33:9459–…
  12. Zongxi Li, Yang Li, Haoran Xie, and S Joe Qin. 2025. Condambigqa: A benchmark and dataset for conditional ambiguous question answering. arXiv preprint arXiv:2502.01523. find
  13. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. find
  14. Sewon Min, Julian Michael, Hannaneh Hajishirzi, and Luke Zettlemoyer. 2020. AmbigQA: Answering ambiguous open-domain questions. In Proceedings of 27098 10 the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 5783– 5797, Online. Association for Computational Linguist… find
  15. Ivan Stelmakh, Yi Luan, Bhuwan Dhingra, and Ming-Wei Chang. 2022. Asqa: Factoid questions meet long-form answers. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 8273–8288.
  16. Anfu Tang, Laure Soulier, and Vincent Guigue. 2025. Clarifying ambiguities: on the role of ambiguity types in prompting methods for clarification generation. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 20–30. find
  17. Yixuan Tang and Yi Yang. 2024. Multihop-rag: Benchmarking retrieval-augmented generation for multi-hop queries. In First Conference on Language Modeling.
  18. Md Mehrab Tanjim, Yeonjun In, Xiang Chen, Victor S Bursztyn, Ryan A Rossi, Sungchul Kim, Guang-Jie Ren, Vaishnavi Muppala, Shun Jiang, Yongsung Kim, and 1 others. 2025. Disambiguation in conversational question answering in the era of llm: A survey. arXiv preprint arXiv:2505.12543. — [arXiv:2505.12543](https://arxiv.org/abs/2505.12543) find
  19. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2022. musique: Multi-hop questions via single-hop question composition. Transactions of the Association for Computational Linguistics, 10:539–554. find
  20. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824– 24837.
  21. Natural Language Processing, EMNLP 2021, pages 7371–7387. Association for Computational Linguistics (ACL). find
  22. Michael JQ Zhang and Eunsol Choi. 2025. Clarify when necessary: Resolving ambiguity through interaction with lms. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 5526–5543. find
  23. Michael JQ Zhang, W Bradley Knox, and Eunsol Choi. 2025a. Modeling future conversation turns to teach llms to ask clarifying questions. In The Thirteenth International Conference on Learning Representations. find
  24. Yanzhao Zhang, Mingxin Li, Dingkun Long, Xin Zhang, Huan Lin, Baosong Yang, Pengjun Xie, An Yang, Dayiheng Liu, Junyang Lin, Fei Huang, and Jingren Zhou. 2025b. Qwen3 embedding: Advancing text embedding and reranking through foundation models. arXiv preprint arXiv:2506.05176.
  25. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Tianle Li, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zhuohan Li, Zi Lin, Eric Xing, Joseph E. Gonzalez, Ion Stoica, and Hao Zhang. 2024. LMSYS-chat-1m: A large-scale real-world LLM conversation dataset. In The Twelfth International Conference on Learning…
  26. Andrew Zhu, Alyssa Hwang, Liam Dugan, and Chris Callison-Burch. 2024. Fanoutqa: A multi-hop, multi-document question answering benchmark for large language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 18–37.
  27. 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. arXiv preprint arXiv:2505.09388.
  28. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William Cohen, Ruslan Salakhutdinov, and Christopher D Manning. 2018. Hotpotqa: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2369–…
  29. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023. React: Synergizing reasoning and acting in language models. In International Conference on Learning Representations (ICLR).
  30. Michael JQ Zhang and Eunsol Choi. 2021. Situatedqa: Incorporating extra-linguistic contexts into qa. In 2021 Conference on Empirical Methods in 27099 11 find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/park-2026-march-evaluating-intersection-ambiguity-interpretation/