Elamur Research⌕
☰
home / papers / 2026 / Moneyball with LLMs: Analyzing Tabular Summarization in Sports Narratives

Moneyball with LLMs: Analyzing Tabular Summarization in Sports Narratives

Paper · 2026 · Ritam Upadhyay, Naman Ahuja, Rishabh Baral et al.

Summary

The paper asks whether LLMs can keep accurate, consistent state about many entities while reading a long, evolving narrative and turning it into statistical tables. Most prior text-to-table work relies on costly prompting or decomposition pipelines and is tested mainly on LiveSum, which says little about how these methods generalize. The authors build SporTabSet, a diagnostic benchmark from two sports. Cricket ball-by-ball commentary from Cricinfo requires separate batsman and bowler scorecards governed by role rules. ESPN basketball play-by-play requires one wide box score covering about 24 players. Using this benchmark, they compare several long-context LLMs under four strategies: plain CoT prompting, decomposition, entity-centric generation, and a symbolic Text–Tuple–Table pipeline. They also test robustness with label-invariant perturbations. Decomposition clearly improves accuracy and numerical fidelity, mostly because it reduces interference between entities rather than improving arithmetic. Models remain sensitive to surface cues and fail in patterned ways: hallucinating entries, omitting players, and confusing roles. The authors conclude that multi-entity memory is the central bottleneck for reliable long-context table generation.

Key points

  • Cricket data: more than 6,500 cricket commentaries, based on 498 ODI innings (2006–2025), with truncated variants of 120, 180, and 240 balls. Gold tables are produced by a regex-based parser that handles extras deterministically.
  • Basketball data: 1,988 games (998 from 2021–2022, 990 from 2025), paired with official box scores of 10 numeric columns. Perturbations expand these to 5,964 evaluation instances.
  • Validation: cricket tables built from full innings match official scorecards 99.7% of the time. Sliced samples (500 checked per setting) reach 99.3% cell-level accuracy and 100% for player names. A manually checked 10% of basketball games shows full alignment between commentary and box score.
  • Trade-offs: the strongest-scoring methods, such as tuple generation over long contexts, carry higher token cost and implementation complexity. Parallel per-entity calls increase input tokens.
  • Release: benchmark artifacts are distributed under CC BY-NC-SA 4.0, for non-commercial research only.

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

🎓Arizona State UniversityTempe · United States50 works · 2,032 citations

 Fields, topics and concepts

📚 Shelves 5The 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 works

🧭 Topics 1Research subjects and languages the work is about.Summarization · 232 works

Cite

Cite as Ritam Upadhyay, Naman Ahuja, Rishabh Baral, Aparna Garimella and Vivek Gupta. 2026. Moneyball with LLMs: Analyzing Tabular Summarization in Sports Narratives. 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.2072

@inproceedings{upadhyay2026moneyball,
  title = {{Moneyball with LLMs: Analyzing Tabular Summarization in Sports Narratives}},
  author = {Ritam Upadhyay, Naman Ahuja, Rishabh Baral, Aparna Garimella and Vivek Gupta},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.2072},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.2072}
}

References

37 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Naman Ahuja, Fenil Bardoliya, Chitta Baral, and Vivek Gupta. 2025. Map&make: Schema guided text to table generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 30249– 30262, Vienna, Austria. Association for Computational… find
  2. Naman Ahuja, Saniya Mulla, Muhammad Ali Khan, Zaryab Bin Riaz, Kaneez Zahra Rubab Khakwani, Mohamad Bassam Sonbol, Irbaz Bin Riaz, and Vivek Gupta. 2026. Evisearch: A human in the loop system for extracting and auditing clinical evidence for systematic reviews. Preprint, arXiv:2604.14165. — [arXiv:2604.14165](https://arxiv.org/abs/2604.14165) find
  3. Tejas Anvekar, Juhna Park, Aparna Garimella, and Vivek Gupta. 2025. Tabrex : Tabular referenceless explainable evaluation. Preprint, arXiv:2512.15907. find
  4. Thomas Ball, Shuo Chen, and Cormac Herley. 2024. Can we count on llms? the fixed-effect fallacy and claims of gpt-4 capabilities. arXiv preprint arXiv:2409.07638. — [arXiv:2409.07638](https://arxiv.org/abs/2409.07638) find
  5. Zhoujun Cheng, Haoyu Dong, Zhiruo Wang, Ran Jia, Jiaqi Guo, Yan Gao, Shi Han, Jian-Guang Lou, and Dongmei Zhang. 2021. Hitab: A hierarchical table dataset for question answering and natural language generation. arXiv preprint arXiv:2108.06712. find
  6. Zheye Deng, Chunkit Chan, Weiqi Wang, Yuxi Sun, Wei Fan, Tianshi Zheng, Yauwai Yim, and Yangqiu Song. 2024a. Text-tuple-table: Towards information integration in text-to-table generation via global tuple extraction. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Proce… find
  7. Zheye Deng, Chunkit Chan, Weiqi Wang, Yuxi Sun, Wei Fan, Tianshi Zheng, Yauwai Yim, and Yangqiu Song. 2024b. Text-Tuple-Table: Towards Information Integration in Text-to-Table Generation via Global Tuple Extraction. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Proce… find
  8. Junjie Fang, Likai Tang, Hongzhe Bi, Yujia Qin, Si Sun, Zhenyu Li, Haolun Li, Yongjian Li, Xin Cong, Yankai Lin, Yukun Yan, Xiaodong Shi, Sen Song, Zhiyuan Liu, and Maosong Sun. 2024. Unimem: Towards a unified view of long-context large language models. In First Conference on Language Modeling. find
  9. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — The Llama 3 Herd of Models
  10. Ruike Hu and Shulei Wu. 2025. Rl-struct: A lightweight reinforcement learning framework for reliable structured output in llms. Preprint, arXiv:2512.00319. — [arXiv:2512.00319](https://arxiv.org/abs/2512.00319) find
  11. Yebowen Hu, Kaiqiang Song, Sangwoo Cho, Xiaoyang Wang, Hassan Foroosh, Dong Yu, and Fei Liu. 2024. Sportsmetrics: Blending text and numerical data to understand information fusion in llms. Preprint, arXiv:2402.10979. find
  12. Wenlong Huang, Fei Xia, Dhruv Shah, Danny Driess, Andy Zeng, Yao Lu, Pete Florence, Igor Mordatch, Sergey Levine, Karol Hausman, and brian ichter. 2023. Grounded decoding: Guiding text generation with grounded models for embodied agents. In Thirty-seventh Conference on Neural Information Processing… find
  13. Parag Jain, Andreea Marzoca, and Francesco Piccinno. 2024. STRUCTSUM generation for faster text comprehension. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7876–7896, Bangkok, Thailand. Association for Computational Linguis… find
  14. Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of hallucination in natural language generation. ACM Comput. Surv., 55(12).
  15. Robin Jia and Percy Liang. 2017. Adversarial examples for evaluating reading comprehension systems. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2275–2285, Copenhagen, Denmark. Association for Computational Linguistics.
  16. Mandar Joshi, Omer Levy, Daniel S. Weld, and Luke Zettlemoyer. 2019. Bert for coreference resolution: Baselines and analysis. ArXiv, abs/1908.09091.
  17. Nikhil Kandpal, Eric Wallace, and Colin Raffel. 2022. Deduplicating training data mitigates privacy risks in language models. ArXiv, abs/2202.06539.
  18. Divyansh Kaushik, Eduard Hovy, and Zachary Lipton. 2020. Learning the difference that makes a difference with counterfactually-augmented data. In International Conference on Learning Representations. find
  19. Tong Li, Zhihao Wang, Liangying Shao, Xuling Zheng, Xiaoli Wang, and Jinsong Su. 2023. A sequence-to-sequence&set model for text-to-table generation. In Findings of the Association for Computational Linguistics: ACL 2023, pages 5358–5370. find
  20. Yiqi Li, Yusheng Liao, Zhe Chen, Yanfeng Wang, and Yu Wang. 2025. DICE: Structured reasoning in LLMs through SLM-guided chain-of-thought correction. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 6951–6966, Suzhou, China. Association for Computationa… find
  21. Nelson F. Liu, Kevin Levonian, Iz Beltagy, Kyle Lo, and Noah A. Smith. 2024. Lost in the middle: How language models use long contexts. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Pa…
  22. R. Thomas McCoy, Ellie Pavlick, and Tal Linzen. 2019. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume…
  23. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 262 others.… — GPT-4 Technical Report
  24. Vihang Pancholi, Jainit Sushil Bafna, Tejas Anvekar, Manish Shrivastava, and Vivek Gupta. 2025. TabX-Eval: Why this is a bad table? an eXhaustive rubric for table evaluation. In Findings of the Association for Computational Linguistics: ACL 2025, pages 22913–22934, Vienna, Austria. Association for C… find
  25. Michał Pietruszka, Michał Turski, Łukasz Borchmann, Tomasz Dwojak, Gabriela Nowakowska, Karolina Szyndler, Dawid Jurkiewicz, and Łukasz Garncarek. 2024. Stable: Table generation framework for encoder-decoder models. In Proceedings of the 18th Conference of the European Chapter of the Association for… find
  26. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, and 24 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15… find
  27. Marco Tulio Ribeiro, Tongshuang Wu, Carlos Guestrin, and Sameer Singh. 2020. CheckList: Beyond accuracy in NLU evaluation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4902–4912, Online. Association for Computational Linguistics. find
  28. Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, and 1 others. 2023. Toolformer: Language models can teach themselves to use tools. In Proceedings of the International Conference on Learning Representations (ICLR).
  29. Anirudh Sundar, Christopher Richardson, and Larry Heck. 2024. gtbls: Generating tables from text by conditional question answering. arXiv preprint arXiv:2403.14457. — [arXiv:2403.14457](https://arxiv.org/abs/2403.14457) find
  30. Sandeep Suntwal, Mithun Paul, Rebecca Sharp, and Mihai Surdeanu. 2019. On the importance of delexicalization for fact verification. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMN… find
  31. Darren Yow-Bang Wang, Zhengyuan Shen, Soumya Smruti Mishra, Zhichao Xu, Yifei Teng, and Haibo Ding. 2025a. Slot: Structuring the output of large language models. Preprint, arXiv:2505.04016. find
  32. Qingyue Wang, Yanhe Fu, Yanan Cao, Shuai Wang, Zhiliang Tian, and Liang Ding. 2025b. Recursively summarizing enables long-term dialogue memory in large language models. Preprint, arXiv:2308.15022. — [arXiv:2308.15022](https://arxiv.org/abs/2308.15022) find
  33. Weizhi Wang, Li Dong, Hao Cheng, Xiaodong Liu, Xifeng Yan, Jianfeng Gao, and Furu Wei. 2023. Augmenting language models with long-term memory. In Thirty-seventh Conference on Neural Information Processing Systems. find
  34. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed H. Chi, Quoc V Le, and Denny Zhou. 2022. Chain of Thought Prompting Elicits Reasoning in Large Language Models. In Advances in Neural Information Processing Systems.
  35. Xueqing Wu, Jiacheng Zhang, and Hang Li. 2022. Text-to-Table: A New Way of Information Extraction. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2518–2533, Dublin, Ireland. Association for Computational Linguistics. find
  36. Shunyu Yao, Jeffrey Zhao, Dian Yu, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. 2022. React: Synergizing reasoning and acting in language models. In NeurIPS 2022 Foundation Models for Decision Making Workshop.
  37. Fengbin Zhu, Ziyang Liu, Fuli Feng, Chao Wang, Moxin Li, and Tat Seng Chua. 2024. Tat-llm: A specialized language model for discrete reasoning over financial tabular and textual data. In Proceedings of the 5th ACM International Conference on AI in Finance, pages 310–318. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/upadhyay-2026-moneyball-llms-analyzing-tabular-summarization/