A Causal Framework to Quantify the Robustness of Mathematical Reasoning with Language Models
Open accessAbstract
Alessandro Stolfo, Zhijing Jin, Kumar Shridhar, Bernhard Schoelkopf, Mrinmaya Sachan. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Stolfo et al. evaluate LLMs on math word problems by measuring sensitivity and robustness to different input interventions and report that models often rely on surface-level cues rather than genuine reasoning.
“Related to this direction, Stolfo et al. (2023) evaluate LLMs on math word problems by measuring sensitivity and robustness to different input interventions and report that models often rely on surface-level cues rather than genuine reasoning, while Hong et al. (2024) developed an ontology and used it to perturb the GSM8K dataset with the aid of LLMs, and Srivastava et al. (2024) purposes function”
From GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models (2024) · cites this paper as Stolfo et al. (2023)
The framework description and §5 Results (page not available) confirm that Stolfo et al. test math word problems using input interventions to measure sensitivity and robustness. Section 5.2 finds that irrelevant textual information affects predictions, particularly for large instruction-tuned models. That supports susceptibility to surface cues, but not the broader claim that models often rely on those cues rather than genuine reasoning; the Abstract attributes the shallow-pattern finding to prior work. (LLM, conf=0.84)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019) · arXiv · doi:10.48550/arxiv.1910.01108 · Cited by 4,653 · see citing works
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208 · see citing works
PaLM: Scaling Language Modeling with Pathways
Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H. W., Sutton, C., Gehrmann, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.02311 · Cited by 2,590 · see citing works
Training Verifiers to Solve Math Word Problems
Cobbe, K., Kosaraju, V., Bavarian, M., Chen, M., Jun, H., Kaiser, L., Plappert, M., Tworek, J., Hilton, J., Nakano, R., et al. (2021) · arXiv · doi:10.48550/arxiv.2110.14168 · Cited by 575 · see citing works
Locating and Editing Factual Associations in GPT
Meng, K., Bau, D., Andonian, A., & Belinkov, Y. (2022) · arXiv · doi:10.48550/arxiv.2202.05262 · Cited by 230 · see citing works
Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond
Feder, A., Keith, K. A., Manzoor, E., Pryzant, R., Sridhar, D., Wood-Doughty, Z., Eisenstein, J., Grimmer, J., Reichart, R., Roberts, M. E., et al. (2022) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00511 · Cited by 220 · see citing works
Generate & Rank: A Multi-task Framework for Math Word Problems
Shen, J., Yin, Y., Li, L., Shang, L., Jiang, X., Zhang, M., & Liu, Q. (2021) · Findings of the Association for Computational Linguistics: EMNLP 2021 · doi:10.18653/v1/2021.findings-emnlp.195 · Cited by 88 · see citing works
Flagged first, then by citations
Research works printed but not linked: 45
- Taylor Berg-Kirkpatrick and Daniel Spokoyny. 2020. An empirical investigation of contextualized number prediction. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4754-4764, Online. Association for Computational Linguistics. · DOI
- Sid Black, Stella Biderman, Eric Hallahan, Quentin An- thony, Leo Gao, Laurence Golding, Horace He, Con- nor Leahy, Kyle McDonell, Jason Phang, et al. 2022. GPT-NeoX-20B: An open-source autoregressive lan- guage model. arXiv preprint arXiv:2204.06745. · arXiv
- Elizabeth M. Brannon. 2005. The independence of language and mathematical reasoning. Proceedings of the National Academy of Sciences, 102(9):3177- 3178. · DOI
- Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel Ziegler, Jeffrey Wu, Clemens Winter, Chris Hesse, Mark Chen, Eric Sigler, Ma- teusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. 2020. Language models are few-shot learners. In Ad- vances in Neural Information Processing Systems, volume 33, pages 1877-1901.
- Amir Feder, Nadav Oved, Uri Shalit, and Roi Reichart. 2021b. CausaLM: Causal model explanation through counterfactual language models. Computational Lin- guistics, 47(2):333-386. · DOI
- Matthew Finlayson, Aaron Mueller, Sebastian Gehrmann, Stuart Shieber, Tal Linzen, and Yonatan Belinkov. 2021. Causal analysis of syntactic agreement mechanisms in neural language models. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 1828-1843, Online. Association for Computational Linguistics. 2, 5, 9 · DOI
- Leo Gao, Stella Biderman, Sid Black, Laurence Gold- ing, Travis Hoppe, Charles Foster, Jason Phang, Ho- race He, Anish Thite, Noa Nabeshima, et al. 2020. The pile: An 800gb dataset of diverse text for lan- guage modeling. arXiv preprint arXiv:2101.00027. · arXiv
- Mor Geva, Ankit Gupta, and Jonathan Berant. 2020. Injecting numerical reasoning skills into language models. In Proceedings of the 58th Annual Meet- ing of the Association for Computational Linguistics, pages 946-958, Online. Association for Computa- tional Linguistics. · DOI
- Zhihua Jin, Xin Jiang, Xingbo Wang, Qun Liu, Yong Wang, Xiaozhe Ren, and Huamin Qu. 2021a. Numgpt: Improving numeracy ability of generative pre-trained models. arXiv preprint arXiv:2109.03137. · arXiv
- Zhijing Jin, Amir Feder, and Kun Zhang. 2022. CausalNLP tutorial: An introduction to causality for natural language processing. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Tutorial Abstracts, pages 17- 22, Abu Dubai, UAE. Association for Computational Linguistics.
- Zhijing Jin, Zeyu Peng, Tejas Vaidhya, Bernhard Schoelkopf, and Rada Mihalcea. 2021b. Mining the cause of political decision-making from social media: A case study of COVID-19 policies across the US states. In Findings of the Association for Compu- tational Linguistics: EMNLP 2021, pages 288-301, Punta Cana, Dominican Republic. Association for Computational Linguistics. · DOI
- Zhijing Jin, Julius von Kügelgen, Jingwei Ni, Tejas Vaidhya, Ayush Kaushal, Mrinmaya Sachan, and Bernhard Schoelkopf. 2021c. Causal direction of data collection matters: Implications of causal and anticausal learning for NLP. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9499-9513, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics. · DOI
- Divyansh Kaushik, Eduard H. Hovy, and Zachary Chase Lipton. 2020. Learning the difference that makes A difference with counterfactually-augmented data. In 8th International Conference on Learning Represen- tations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.
- Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yu- taka Matsuo, and Yusuke Iwasawa. 2022. Large lan- guage models are zero-shot reasoners. arXiv preprint arXiv:2205.11916. · arXiv
- Rik Koncel-Kedziorski, Subhro Roy, Aida Amini, Nate Kushman, and Hannaneh Hajishirzi. 2016. MAWPS: A math word problem repository. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Hu- man Language Technologies, pages 1152-1157, San Diego, California. Association for Computational Linguistics. 5, 6 · DOI
- Zhiheng Lyu, Zhijing Jin, Justus Mattern, Rada Mi- halcea, Mrinmaya Sachan, and Bernhard Schölkopf. 2023. Psychologically-inspired causal prompts. CoRR, abs/2305.01764. · DOI
- Shen-yun Miao, Chao-Chun Liang, and Keh-Yih Su. 2020. A diverse corpus for evaluating and developing English math word problem solvers. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 975-984, Online. Association for Computational Linguistics. 5, 6 · DOI · Paper page
- Swaroop Mishra, Matthew Finlayson, Pan Lu, Leonard Tang, Sean Welleck, Chitta Baral, Tanmay Rajpuro- hit, Oyvind Tafjord, Ashish Sabharwal, Peter Clark, et al. 2022a. Lila: A unified benchmark for mathe- matical reasoning. arXiv preprint arXiv:2210.17517. · arXiv
- Swaroop Mishra, Arindam Mitra, Neeraj Varshney, Bhavdeep Sachdeva, Peter Clark, Chitta Baral, and Ashwin Kalyan. 2022b. NumGLUE: A suite of fun- damental yet challenging mathematical reasoning tasks. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Vol- ume 1: Long Papers), pages 3505-3523, Dublin, Ireland. Association for Computational Linguistics. · DOI
- Martin M Monti, Lawrence M Parsons, and Daniel N Osherson. 2012. Thought beyond language: Neural dissociation of algebra and natural language. Psycho- logical science, 23(8):914-922. · DOI · Paper page
- Jingwei Ni, Zhijing Jin, Markus Freitag, Mrinmaya Sachan, and Bernhard Schölkopf. 2022. Original or translated? A causal analysis of the impact of translationese on machine translation performance. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computa- tional Linguistics: Human Language Technologies, pages 5303-5320, Seattle, United States. Association for Computational Linguistics. · DOI
- Kuntal Kumar Pal and Chitta Baral. 2021. Investigating numeracy learning ability of a text-to-text transfer model. In Findings of the Association for Computa- tional Linguistics: EMNLP 2021, pages 3095-3101, Punta Cana, Dominican Republic. Association for Computational Linguistics. · DOI
- Arkil Patel, Satwik Bhattamishra, and Navin Goyal. 2021. Are NLP models really able to solve simple math word problems? In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2080-2094, Online. Association for Computational Linguistics. 1, 2, 5, 6, 14 · DOI · Paper page
- Judea Pearl. 1995. Causal diagrams for empirical re- search. Biometrika, 82(4):669-688. 2, 3, 4
- Judea Pearl. 2001. Direct and indirect effects. In UAI '01: Proceedings of the 17th Conference in Uncer- tainty in Artificial Intelligence, University of Wash- ington, Seattle, Washington, USA, August 2-5, 2001, pages 411-420.
- Morgan Kaufmann. 2, 4 Judea Pearl. 2009. Causality. Cambridge University Press.
- Jonas Peters, Dominik Janzing, and Bernhard Schölkopf. 2017. Elements of causal inference: Foundations and learning algorithms. The MIT Press.
- Piotr Piękos, Mateusz Malinowski, and Henryk Michalewski. 2021. Measuring and improving BERT's mathematical abilities by predicting the or- der of reasoning. In Proceedings of the 59th An- nual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer- ence on Natural Language Processing (Volume 2: Short Papers), pages 383-394, Online. Association for Computational Linguistics. · DOI
- Yasaman Razeghi, Robert L Logan IV, Matt Gardner, and Sameer Singh. 2022. Impact of pretraining term frequencies on few-shot reasoning. arXiv preprint arXiv:2202.07206. 1, 9 · arXiv
- Margaret E Roberts, Brandon M Stewart, and Richard A Nielsen. 2020. Adjusting for confounding with text matching. American Journal of Political Science, 64(4):887-903.
- Mrinmaya Sachan, Kumar Dubey, and Eric Xing. 2017. From textbooks to knowledge: A case study in har- vesting axiomatic knowledge from textbooks to solve geometry problems. In Proceedings of the 2017 Con- ference on Empirical Methods in Natural Language Processing, pages 773-784.
- Mrinmaya Sachan, Kumar Avinava Dubey, Tom M Mitchell, Dan Roth, and Eric P Xing. 2018. Learning pipelines with limited data and domain knowledge: A study in parsing physics problems. Advances in Neural Information Processing Systems, 31. 1
- Mrinmaya Sachan and Eric Xing. 2017. Learning to solve geometry problems from natural language demonstrations in textbooks. In Proceedings of the 6th Joint Conference on Lexical and Computational Semantics (* SEM 2017), pages 251-261.
- Minjoon Seo, Hannaneh Hajishirzi, Ali Farhadi, Oren Etzioni, and Clint Malcolm. 2015. Solving geome- try problems: Combining text and diagram interpre- tation. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1466-1476, Lisbon, Portugal. Association for Computational Linguistics. · DOI
- Daniel Spokoyny, Ivan Lee, Zhao Jin, and Taylor Berg- Kirkpatrick. 2021. Masked measurement prediction: Learning to jointly predict quantities and units from textual context. arXiv preprint arXiv:2112.08616. · arXiv
- Avijit Thawani, Jay Pujara, Filip Ilievski, and Pedro Szekely. 2021. Representing numbers in NLP: a survey and a vision. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 644-656, Online. As- sociation for Computational Linguistics. · DOI
- Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and effi- cient foundation language models. arXiv preprint arXiv:2302.13971. 2, 8 · arXiv
- Victor Veitch, Dhanya Sridhar, and David M. Blei. 2020. Adapting text embeddings for causal inference. In Proceedings of the Thirty-Sixth Conference on Un- certainty in Artificial Intelligence, UAI 2020, virtual online, August 3-6, 2020, volume 124 of Proceed- ings of Machine Learning Research, pages 919-928. AUAI Press.
- Jesse Vig, Sebastian Gehrmann, Yonatan Belinkov, Sharon Qian, Daniel Nevo, Yaron Singer, and Stuart Shieber. 2020. Investigating gender bias in language models using causal mediation analysis. Advances in Neural Information Processing Systems, 33:12388- 12401. 2, 9
- Eric Wallace, Yizhong Wang, Sujian Li, Sameer Singh, and Matt Gardner. 2019. Do NLP models know num- bers? probing numeracy in embeddings. In Proceed- ings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th Inter- national Joint Conference on Natural Language Pro- cessing (EMNLP-IJCNLP), pages 5307-5315, Hong Kong, China. Association for Computational Linguis- tics. · DOI · Paper page
- Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, et al. 2022a. Emergent abilities of large language models. arXiv preprint arXiv:2206.07682. 1, 8 · arXiv
- Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed H. Chi, Quoc Le, and Denny Zhou. 2022b. Chain of thought prompting elicits reasoning in large language models. CoRR, abs/2201.11903. · arXiv
- Sean Welleck, Peter West, Jize Cao, and Yejin Choi. 2022. Symbolic brittleness in sequence models: on systematic generalization in symbolic mathematics. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 8629-8637.
- Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, R'emi Louf, Morgan Funtowicz, and Jamie Brew. 2019. HuggingFace's transformers: State-of-the-art natural language processing. arXiv preprint arXiv:1910.03771. · arXiv
- Xikun Zhang, Deepak Ramachandran, Ian Tenney, Yanai Elazar, and Dan Roth. 2020. Do language embeddings capture scales? In Findings of the Asso- ciation for Computational Linguistics: EMNLP 2020, pages 4889-4896, Online. Association for Computa- tional Linguistics. 9 · DOI
Cited by
Sorted by citations
Language Model Behavior: A Comprehensive Survey
Chang, T. A., & Bergen, B. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00492 · Cited by 105 · see citing works
Distilling Reasoning Capabilities into Smaller Language Models
Shridhar, K., Stolfo, A., & Sachan, M. (2023) · Findings of the Association for Computational Linguistics: ACL 2023 · doi:10.18653/v1/2023.findings-acl.441 · Cited by 79 · see citing works
Self-citationGSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
Mirzadeh, I., Alizadeh, K., Shahrokhi, H., Tuzel, O., Bengio, S., & Farajtabar, M. (2024) · arXiv · doi:10.48550/arxiv.2410.05229 · Cited by 57 · see citing works
Unveiling the Key Factors for Distilling Chain-of-Thought Reasoning
Chen, X., Sun, Z., Wenjin, G., Zhang, M., Chen, Y., Sun, Y., Su, H., Pan, Y., Klakow, D., Li, W., et al. (2025) · Findings of the Association for Computational Linguistics: ACL 2025 · doi:10.18653/v1/2025.findings-acl.782 · Cited by 7 · see citing works
- Authors
- Stolfo, A., Jin, Z., Shridhar, K., Schoelkopf, B., & Sachan, M.
- Affiliations
- ETH Zurich (CH)
- Venue
- Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Published
- 2023
- Pages
- 545-561
- ISSN
- 0736-587X (ISSN-L)
- OpenAlex
- W4385565015
- Cited by
- 17
- References
- 7
- Citations by year
- 2026: 1 · 2025: 5 · 2024: 3 · 2023: 7 · 2022: 1
- FWCI
- 2.83
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Proceedings of the conference - Association for Computational Linguistics. Meeting (publishedVersion, cc-by)
- Funders
- John Templeton Foundation (61156)
- Open Philanthropy Project
- Future of Life Institute
- Bundesministerium für Bildung und Forschung (01IS18039B, FKZ: 01IS18039B)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Logic, Reasoning, and Knowledge
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- mathematical reasoning, language models, robustness, causal framework, robustness quantification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
