The DOI matches its Crossref record; 30 of the 44 references it cites were found in a registry or an index.
FinQA: A Dataset of Numerical Reasoning over Financial Data
Open accessAbstract
Zhiyu Chen, Wenhu Chen, Charese Smiley, Sameena Shah, Iana Borova, Dylan Langdon, Reema Moussa, Matt Beane, Ting-Hao Huang, Bryan Routledge, William Yang Wang. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.
Overview
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
Verification
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Most existing financial benchmarks focus on static question-answering, failing to capture the dynamics of real-market trading.”
From StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets? (2025) · cites this paper as Chen et al., 2021
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“However, most existing benchmarks focus on static question-answering, such as FinQA, TAT-QA, and FinBench (Chen et al., 2021; Zhu et al., 2021; Yin et al., 2023).”
From StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets? (2025) · cites this paper as Chen et al., 2021
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“For all models, we use the Adam optimizer (Kingma and Ba, 2015).”
MethodBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 6×
“Retriever The full page of the financial report can go beyond 2,000 tokens, which cannot be coped with the current popular QA models (Devlin et al., 2019).”
BackgroundMethodInfluentialRoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515 · cited 4×
“, 2019) and RoBERTa (Liu et al., 2019), our proposed approach outperforms all other baselines and achieves an execution accuracy of 65.”
BackgroundMethodInfluentialLongformer: The Long-Document Transformer
Beltagy, I., Peters, M. E., & Cohan, A. (2020) · arXiv · doi:10.48550/arxiv.2004.05150 · Cited by 2,281 · cited 6×
“3703 scales linearly with sequence length, like the Longformer (Beltagy et al., 2020).”
BackgroundInfluentialKnow What You Don’t Know: Unanswerable Questions for SQuAD
Rajpurkar, P., Jia, R., & Liang, P. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/p18-2124 · Cited by 2,240 · cited 1×
“Financial QA is more challenging than classic QA (Rajpurkar et al., 2018; Yang et al., 2018) because it requires the system to spot relevant information across heterogeneous sources, such as tables and unstructured texts, and then create a numerical reasoning path to connect all the information.”
BackgroundHotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
Yang, Z., Qi, P., Zhang, S., Bengio, Y., Cohen, W., Salakhutdinov, R., & Manning, C. D. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1259 · Cited by 1,804 · cited 2×
“Financial QA is more challenging than classic QA (Rajpurkar et al., 2018; Yang et al., 2018) because it requires the system to spot relevant information across heterogeneous sources, such as tables and unstructured texts, and then create a numerical reasoning path to connect all the information.”
BackgroundYu, T., Zhang, R., Yang, K., Yasunaga, M., Wang, D., Li, Z., Ma, J., Li, I., Yao, Q., Roman, S., et al. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1425 · Cited by 1,034
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Amini, A., Gabriel, S., Lin, S., Koncel-Kedziorski, R., Choi, Y., & Hajishirzi, H. (2019) · North American Chapter of the Association for Computational Linguistics · doi:10.18653/v1/n19-1245 · Cited by 956 · cited 6×
“This way the model is able to learn the program structures as patterns from very large-scale data (~40k for MathQA), but may fail on learning the reasoning paths.”
BackgroundMethodInfluentialFinBERT: Financial Sentiment Analysis with Pre-trained Language Models
Araci, D. (2019) · arXiv · Cited by 602 · cited 5×
“The performance of using FinBert is no better than BERT-large, mostly because its pre-training corpus is limited (~30M words from news articles).”
BackgroundInfluentialMAWPS: A Math Word Problem Repository
Koncel-Kedziorski, R., Roy, S., Amini, A., Kushman, N., & Hajishirzi, H. (2016) · North American Chapter of the Association for Computational Linguistics · doi:10.18653/v1/n16-1136 · Cited by 551 · cited 2×
“Another kind of QA datasets related to ours is the math word problem datasets, like MaWPS (Koncel-Kedziorski et al., 2016), MathQA (Amini et al.”
BackgroundPoor Numbers: How we are misled by African development statistics and what to do about it
Mesplé-Somps, S. (2014) · African Affairs · doi:10.1093/afraf/adt082 · Cited by 538 · cited 1×
“Financial analysis is a critical means of assessing business performance, and the consequences of poor analysis can involve costs of billions of dollars (Jerven, 2013; MacKenzie, 2008).”
BackgroundThe Web as a Knowledge-Base for Answering Complex Questions
Talmor, A., & Berant, J. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1059 · Cited by 520 · cited 2×
“Popular datasets include ComplexWebQuestions (Talmor and Berant, 2018), WikiTableQuestions (Pasupat and Liang, 2015), Spider (Yu et al.”
Compositional Semantic Parsing on Semi-Structured Tables
Pasupat, P., & Liang, P. (2015) · Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.3115/v1/p15-1142 · Cited by 514 · cited 2×
“Popular datasets include ComplexWebQuestions (Talmor and Berant, 2018), WikiTableQuestions (Pasupat and Liang, 2015), Spider (Yu et al.”
FinBERT: A Pretrained Language Model for Financial CommunicationsPreprint
Yang, Y., Mark Christopher Siy, U. Y., & Huang, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.08097 · Cited by 484 · cited 1×
“Recent works attempt to develop pre-trained models specialized for finance domain (Yang et al., 2020; Araci, 2019), and the downstream tasks are mostly sentiment classifications.”
BackgroundBernard Mandeville and the 'Economy' of the Dutch
Bick, A. (2008) · doi:10.23941/ejpe.v1i1.9 · Cited by 426 · cited 1×
“Financial analysis is a critical means of assessing business performance, and the consequences of poor analysis can involve costs of billions of dollars (Jerven, 2013; MacKenzie, 2008).”
BackgroundDROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
Dua, D., Wang, Y., Dasigi, P., Stanovsky, G., Singh, S., & Gardner, M. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1246 · Cited by 393 · cited 7×
“As most existing QA datasets were constructed by MTurk workers (Yang et al., 2018; Dua et al., 2019; Chen et al., 2020c), it requires substantial domain-specific knowledge to compose meaningful questions that are hard for computers to answer.”
BackgroundMethodInfluentialPoor numbers: how we are misled by African development statistics and what to do about it
Berry, S. (2013) · doi:10.1080/10796126.2013.833894 · Cited by 335 · cited 1×
“Financial analysis is a critical means of assessing business performance, and the consequences of poor analysis can involve costs of billions of dollars (Jerven, 2013; MacKenzie, 2008).”
BackgroundHybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data
Chen, W., Zha, H., Chen, Z., Xiong, W., Wang, H., & Wang, W. Y. (2020) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2020.findings-emnlp.91 · Cited by 290 · cited 6×
“As most existing QA datasets were constructed by MTurk workers (Yang et al., 2018; Dua et al., 2019; Chen et al., 2020c), it requires substantial domain-specific knowledge to compose meaningful questions that are hard for computers to answer.”
BackgroundMethodInfluentialSelf-citationZheng, X., Burdick, D., Popa, L., Zhong, X., & Wang, N. (2020) · IEEE Workshop/Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00074 · Cited by 248 · cited 2×
“We develop FINQA based on the publicly available earnings reports of S&P 500 companies from 1999 to 2019, collected in the FinTabNet dataset (Zheng et al., 2021).”
BackgroundMethodTabFact: A Large-scale Dataset for Table-based Fact Verification
Chen, W., Wang, H., Chen, J., Zhang, Y., Wang, H., Li, S., Zhou, X., & Wang, W. Y. (2019) · arXiv · doi:10.48550/arxiv.1909.02164 · Cited by 239 · cited 6×
“As most existing QA datasets were constructed by MTurk workers (Yang et al., 2018; Dua et al., 2019; Chen et al., 2020c), it requires substantial domain-specific knowledge to compose meaningful questions that are hard for computers to answer.”
BackgroundMethodInfluentialSelf-citationDeep learning for financial sentiment analysis on finance news providers
Day, M.-Y., & Lee, C.-C. (2016) · International Conference on Advances in Social Networks Analysis and Mining · doi:10.1109/asonam.2016.7752381 · Cited by 155
A BERT Baseline for the Natural QuestionsPreprint
Alberti, C., Lee, K., & Collins, M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.08634 · Cited by 154 · cited 1×
“Another common strategy is sliding window (Alberti et al., 2019).”
Data-Driven Market-Making via Model-Free Learning
Zhong, Y., Bergstrom, Y., & Ward, A. R. (2020) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/615 · Cited by 132 · cited 4×
“A few NLP studies in Question Answering (QA) explored the numerical reasoning capabilities needed to answer questions correctly.”
BackgroundMethodInfluentialA Multilayer Perceptron based Ensemble Technique for Fine-grained Financial Sentiment Analysis
Akhtar, M. S., Abhishek, K., Ghosal, D., Ekbal, A., & Bhattacharyya, P. (2017) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d17-1057 · Cited by 123 · cited 1×
“…detect fraud (Han et al., 2018; Wang et al., 2019; Nourbakhsh and Bang, 2019), sentiment analysis to assist market prediction (Day and Lee, 2016; Wang et al., 2013; Akhtar et al., 2017), opinionated Question Answering (Liu et al., 2020), such as the FiQA2 dataset built from forums and social media.”
MethodLange, A.-C., Lenglet, M., & Seyfert, R. (2016) · Economy and Society · doi:10.1080/03085147.2016.1213986 · Cited by 78 · cited 2×
“com/czyssrs/FinQA by an exponentially expanding collection of company financial documents (MacKenzie et al., 2012; Lange et al., 2016) such that it is genuinely unclear whether dedicated human effort can produce fiscal analysis of sufficient quality for current decision making.”
BackgroundHan, J., Barman, U., Hayes, J., Du, J., Burgin, E., & Wan, D. (2018) · Annual Meeting of the Association for Computational Linguistics: System Demonstrations (ACL Demo) · doi:10.18653/v1/p18-4007 · Cited by 53 · cited 2×
“Previous works in finance domain include risk management to detect fraud (Han et al., 2018; Wang et al., 2019; Nourbakhsh and Bang, 2019), sentiment analysis to assist market prediction (Day and Lee, 2016; Wang et al., 2013; Akhtar et al., 2017), opinionated Question Answering (Liu et al., 2020),…”
BackgroundKim, B., Ki, K. S., Lee, D., & Gweon, G. (2020) · Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2020.emnlp-main.308 · Cited by 45 · cited 1×
“Existing models include (Kim et al., 2020; Chen et al., 2020a,d), etc. Financial NLP. Financial NLP has become one of the major application domains attracting growing attentions.”
MethodAre You for Real? Detecting Identity Fraud via Dialogue Interactions
Wang, W.-K., Zhang, J., Li, Q., Zong, C.-Q., & Li, Z. (2019) · Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d19-1185 · Cited by 19 · cited 2×
“Previous works in finance domain include risk management to detect fraud (Han et al., 2018; Wang et al., 2019; Nourbakhsh and Bang, 2019), sentiment analysis to assist market prediction (Day and Lee, 2016; Wang et al., 2013; Akhtar et al., 2017), opinionated Question Answering (Liu et al., 2020),…”
BackgroundSelf-citationEllsmore, J. (2011) · doi:10.7551/mitpress/11819.003.0003 · Cited by 16
A framework for anomaly detection using language modeling, and its applications to financePreprint
Nourbakhsh, A., & Bang, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.09156 · Cited by 15 · cited 1×
“Previous works in finance domain include risk management to detect fraud (Han et al., 2018; Wang et al., 2019; Nourbakhsh and Bang, 2019), sentiment analysis to assist market prediction (Day and Lee, 2016; Wang et al.”
Background
Research works printed but not linked: 7
- additionally , we have other committed and uncommitted credit lines of $ 746 million with major international banks and financial institutions to support our general global funding needs , including with respect to bank supported letters of credit, performance bonds and guarantees .
- Kezhen Chen, Qiuyuan Huang, Hamid Palangi, Paul Smolensky, Kenneth D. Forbus, and Jianfeng Gao. 2020a. Mapping natural-language problems to formal-language solutions using structured neural representations. In Proceedings of the 37th Inter- national Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event, volume 119 of Proceedings of Machine Learning Research, pages 1566-1575. PMLR.
- Xinyun Chen, Chen Liang, Adams Wei Yu, Denny Zhou, Dawn Song, and Quoc V. Le. 2020d. Neu- ral symbolic reader: Scalable integration of dis- tributed and symbolic representations for reading comprehension. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.
- Zhuang Liu, Degen Huang, Kaiyu Huang, Zhuang Li, and Jun Zhao. 2020. Finbert: A pre-trained finan- cial language representation model for financial text mining. In Proceedings of the Twenty-Ninth Inter- national Joint Conference on Artificial Intelligence, IJCAI 2020, pages 4513-4519. ijcai.org. · DOI
- Donald MacKenzie. 2008. An engine, not a camera: How financial models shape markets. Mit Press.
- Donald MacKenzie, Daniel Beunza, Yuval Millo, and Juan Pablo Pardo-Guerra. 2012. Drilling through the allegheny mountains: Liquidity, materiality and high-frequency trading. Journal of cultural econ- omy, 5(3):279-296.
- Chuan-Ju Wang, Ming-Feng Tsai, Tse Liu, and Chin- Ting Chang. 2013. Financial sentiment analysis for risk prediction. In Sixth International Joint Conference on Natural Language Processing, IJC- NLP 2013, Nagoya, Japan, October 14-18, 2013, pages 802-808. Asian Federation of Natural Lan- guage Processing / ACL.
Cited by
Sorted by citations
A Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4Preprint
Kalyan, K. S. (2023) · Natural Language Processing Journal · doi:10.2139/ssrn.4593895 · Cited by 438
What matters when building vision-language models?Preprint
Laurençon, H., Tronchon, L., Cord, M., & Sanh, V. (2024) · Neural Information Processing Systems · doi:10.48550/arxiv.2405.02246 · Cited by 413
“TabMWP (Lu et al., 2023) 22,729 23,059 1,948,166 2.49% TAT-QA (Zhu et al., 2021) 2,199 13,215 283,776 2.18% HiTab (Cheng et al., 2022) 2,500 7,782 351,299 1.80% MultiHiertt (Zhao et al., 2022) 7,619 7,830 267,615 1.37% FinQA (Chen et al., 2021) 5,276 6,251 242,561 0.99% WikiSQL (Zhong et al., 2017) 74,989 86,202 9,680…”
InfluentialBloombergGPT: A Large Language Model for Finance
Wu, S., Irsoy, O., Lu, S., Dabravolski, V., Dredze, M., Gehrmann, S., Kambadur, P., Rosenberg, D., & Mann, G. (2023) · arXiv · Cited by 357
“Financial NLP tasks (Shah et al., 2022) include sentiment analysis (Araci, 2019), named entity recognition (Salinas Alvarado et al., 2015), news classification (Sinha and Khandait, 2020), and question answering (Chen et al., 2021b, 2022).”
BackgroundPIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for FinancePreprint
Xie, Q., Han, W., Xiao, Z., Lai, Y., Peng, M., Lopez-Lira, A., & Huang, J. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.05443 · Cited by 294
“ConvFinQA has multi-turn conversations with questions and answering.”
MethodInfluentialA Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4
Kalyan, K. S. (2024) · Natural Language Processing Journal · doi:10.1016/j.nlp.2023.100048 · Cited by 335
Metadata
- Authors
- Chen, Z., Chen, W., Smiley, C., Shah, S., Borova, I., Langdon, D., Moussa, R., Beane, M., Huang, T.-H., Routledge, B., & Wang, W. Y.
- Affiliations
- University of California, Santa Barbara (US)
- Thomson Reuters (United States) (US)
- Pennsylvania State University (US)
- Carnegie Mellon University (US)
- Stanford University (US)
- ORCID
- Venue
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Published
- 2021
- Pages
- 3697-3711
- OpenAlex
- W3198243861
- Cited by
- 258
- References
- 44
- Citations by year
- 2026: 79 · 2025: 74 · 2024: 42 · 2023: 35 · 2022: 27 · 2021: 1
- FWCI
- 12.13
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- numerical reasoning, financial data, question answering, financial question answering
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
