Elamur Research
☰
home / papers / 2023 / ToolQA: A Dataset for LLM Question Answering with External Tools

ToolQA: A Dataset for LLM Question Answering with External Tools

Written for the Research library (claude-opus-5-5) from the paper text. Original — arxiv.org/pdf/2306.13304

Summary

LLMs often hallucinate and struggle with arithmetic, and connecting them to external tools can help. Existing benchmarks, however, cannot tell whether a model actually used a tool or simply recalled facts it had seen during pre-training. ToolQA addresses this by building questions that can be solved only by querying external reference corpora (text, tables and graphs) chosen to have little overlap with pre-training data. Construction is largely automated in three stages. First, suitable public corpora are collected. Next, LLMs propose question templates under human guidance, and the templates are validated and instantiated. Finally, answers are computed by programs that implement the tool operations, so correctness does not depend on manual annotation. The benchmark is split into easy and hard questions. Models relying only on internal knowledge almost always fail. Tool-augmented agents such as ReAct and Chameleon do better, but they remain weak on hard, multi-tool questions. The resulting testbed separates genuine tool use from memorization.

Key points

  • The benchmark spans 8 domains and offers 13 tools covering retrieval, database operations, SQL/code interpretation, math and graph queries. Example sources include flight records, coffee prices, Yelp, Agenda, SciREX and DBLP.
  • ChatGPT and Chain-of-thought prompting score about 5% on easy questions and about 2% on hard ones.
  • The best tool-augmented method reaches 43.15% on easy questions but only 8.2% on hard questions.
  • A manual error analysis of ReAct (GPT-3.5) found argument errors to be the most common failure (48.23% easy, 44.56% hard). Querying the wrong data source rises sharply on hard questions (18.04% vs. 7.48%). The analysis also notes "innovative" tool combinations that are frequently accompanied by hallucinated observations.
  • Data and code are publicly available at github.com/night-chen/ToolQA.

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

Fields: Natural language processing · AI & machine learning

Cited by

118 papers of our library cite it (of the 140673 whose bibliographies we read); 8 of them have a page here, 110 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2306.13304 · arXiv 2306.13304

@misc{zhuang2023toolqa,
  title = {{ToolQA: A Dataset for LLM Question Answering with External Tools}},
  author = {Zhuang, Yuchen and Yu, Yue and Wang, Kuan and Sun, Haotian and Zhang, Chao},
  year = {2023},
  doi = {10.48550/arxiv.2306.13304},
  eprint = {2306.13304},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2306.13304}
}

References

70 references, 34 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Y. Bai, A. Jones, K. Ndousse, A. Askell, A. Chen, N. DasSarma, D. Drain, S. Fort, D. Ganguli, T. Henighan, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862, 2022. find
  2. S. Borgeaud, A. Mensch, J. Hoffmann, T. Cai, E. Rutherford, K. Millican, G. B. Van Den Driessche, J.-B. Lespiau, B. Damoc, A. Clark, et al. Improving language models by retrieving from trillions of tokens. In International conference on machine learning, pages 2206–2240. PMLR, 2022. find
  3. T. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell, et al. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020. find
  4. S. Bubeck, V. Chandrasekaran, R. Eldan, J. Gehrke, E. Horvitz, E. Kamar, P. Lee, Y. T. Lee, Y. Li, S. Lundberg, et al. Sparks of artificial general intelligence: Early experiments with gpt-4. arXiv preprint arXiv:2303.12712, 2023. find our card
  5. W. Chen, X. Ma, X. Wang, and W. W. Cohen. Program of thoughts prompting: Disentangling computation from reasoning for numerical reasoning tasks, 2022. find our card
  6. W. Chen, X. Wang, and W. Y. Wang. A dataset for answering time-sensitive questions. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021. find
  7. A. Chowdhery, S. Narang, J. Devlin, M. Bosma, G. Mishra, A. Roberts, P. Barham, H. W. Chung, C. Sutton, S. Gehrmann, et al. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:2204.02311, 2022. find
  8. H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, E. Li, X. Wang, M. Dehghani, S. Brahma, et al. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416, 2022. find
  9. K. Cobbe, V. Kosaraju, M. Bavarian, M. Chen, H. Jun, L. Kaiser, M. Plappert, J. Tworek, J. Hilton, R. Nakano, et al. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168, 2021. find our card
  10. B. Dhingra, J. R. Cole, J. M. Eisenschlos, D. Gillick, J. Eisenstein, and W. W. Cohen. Time-aware language models as temporal knowledge bases. Transactions of the Association for Computational Linguistics, 10:257–273, 2022. find
  11. L. Gao, A. Madaan, S. Zhou, U. Alon, P. Liu, Y. Yang, J. Callan, and G. Neubig. Pal: Program-aided language models. arXiv preprint arXiv:2211.10435, 2022. find
  12. D. Hendrycks, C. Burns, S. Kadavath, A. Arora, S. Basart, E. Tang, D. Song, and J. Steinhardt. Measuring mathematical problem solving with the math dataset. NeurIPS, 2021. find our card
  13. S. Imani, L. Du, and H. Shrivastava. Mathprompter: Mathematical reasoning using large language models. arXiv preprint arXiv:2303.05398, 2023. find our card
  14. G. Izacard, M. Caron, L. Hosseini, S. Riedel, P. Bojanowski, A. Joulin, and E. Grave. Towards unsupervised dense information retrieval with contrastive learning. arXiv preprint arXiv:2112.09118, 2021. find our card
  15. G. Izacard, P. Lewis, M. Lomeli, L. Hosseini, F. Petroni, T. Schick, J. Dwivedi-Yu, A. Joulin, S. Riedel, and E. Grave. Few-shot learning with retrieval augmented language models. arXiv preprint arXiv:2208.03299, 2022. find
  16. S. Jain, M. van Zuylen, H. Hajishirzi, and I. Beltagy. SciREX: A challenge dataset for document-level information extraction. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7506–7516, Online, July 2020. Association for Computational Linguistics. find
  17. Z. Ji, N. Lee, R. Frieske, T. Yu, D. Su, Y. Xu, E. Ishii, Y. J. Bang, A. Madotto, and P. Fung. Survey of hallucination in natural language generation. ACM Computing Surveys, 55(12):1–38, 2023. find our card
  18. Q. Jin, Y. Yang, Q. Chen, and Z. Lu. Genegpt: Augmenting large language models with domain tools for improved access to biomedical information. ArXiv, 2023. find
  19. M. Joshi, E. Choi, D. Weld, and L. Zettlemoyer. TriviaQA: A large scale distantly supervised challenge dataset for reading comprehension. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1601–1611, Vancouver, Canada, July 2017.… find our card
  20. V. Karpukhin, B. Oguz, S. Min, P. Lewis, L. Wu, S. Edunov, D. Chen, and W.-t. Yih. Dense passage retrieval for open-domain question answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6769–6781, 2020. find our card
  21. J. Kasai, K. Sakaguchi, Y. Takahashi, R. L. Bras, A. Asai, X. Yu, D. Radev, N. A. Smith, Y. Choi, and K. Inui. Realtime qa: What’s the answer right now? arXiv preprint arXiv:2207.13332, 2022. find our card
  22. G. Kim, P. Baldi, and S. McAleer. Language models can solve computer tasks, 2023. find our card
  23. T. Kojima, S. S. Gu, M. Reid, Y. Matsuo, and Y. Iwasawa. Large language models are zero-shot reasoners. In A. H. Oh, A. Agarwal, D. Belgrave, and K. Cho, editors, Advances in Neural Information Processing Systems, 2022. find our card
  24. P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Küttler, M. Lewis, W.-t. Yih, T. Rocktäschel, et al. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems, 33:9459–9474, 2020. find
  25. A. Lewkowycz, A. Andreassen, D. Dohan, E. Dyer, H. Michalewski, V. Ramasesh, A. Slone, C. Anil, I. Schlag, T. Gutman-Solo, et al. Solving quantitative reasoning problems with language models. arXiv preprint arXiv:2206.14858, 2022. find our card
  26. M. Li, F. Song, B. Yu, H. Yu, Z. Li, F. Huang, and Y. Li. Api-bank: A benchmark for tool-augmented llms, 2023. find
  27. B. Y. Lin, K. Tan, C. S. Miller, B. Tian, and X. Ren. Unsupervised cross-task generalization via retrieval augmentation. In Advances in Neural Information Processing Systems, 2022. find
  28. P. Lu, B. Peng, H. Cheng, M. Galley, K.-W. Chang, Y. N. Wu, S.-C. Zhu, and J. Gao. Chameleon: Plug-and-play compositional reasoning with large language models. arXiv preprint arXiv:2304.09842, 2023. find our card
  29. P. Lu, L. Qiu, K.-W. Chang, Y. N. Wu, S.-C. Zhu, T. Rajpurohit, P. Clark, and A. Kalyan. Dynamic prompt learning via policy gradient for semi-structured mathematical reasoning. arXiv preprint arXiv:2209.14610, 2022. find our card
  30. S. Lu, N. Duan, H. Han, D. Guo, S.-w. Hwang, and A. Svyatkovskiy. Reacc: A retrieval-augmented code completion framework. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6227–6240, 2022. find
  31. A. Madaan and A. Yazdanbakhsh. Text and patterns: For effective chain of thought, it takes two to tango. arXiv preprint arXiv:2209.07686, 2022. find
  32. A. Mallen, A. Asai, V. Zhong, R. Das, H. Hajishirzi, and D. Khashabi. When not to trust language models: Investigating effectiveness and limitations of parametric and non-parametric memories. arXiv preprint arXiv:2212.10511, 2022. find
  33. S. Mishra, M. Finlayson, P. Lu, L. Tang, S. Welleck, C. Baral, T. Rajpurohit, O. Tafjord, A. Sabharwal, P. Clark, et al. Lila: A unified benchmark for mathematical reasoning. arXiv preprint arXiv:2210.17517, 2022. find our card
  34. R. Nakano, J. Hilton, S. Balaji, J. Wu, L. Ouyang, C. Kim, C. Hesse, S. Jain, V. Kosaraju, W. Saunders, et al. Webgpt: Browser-assisted question-answering with human feedback. arXiv preprint arXiv:2112.09332, 2021. find our card
  35. R. Nogueira, Z. Jiang, and J. Lin. Investigating the limitations of transformers with simple arithmetic tasks. arXiv preprint arXiv:2102.13019, 2021. find
  36. OpenAI. Gpt-4 technical report. arXiv, 2023. find our card
  37. OpenAI. Introducing chatgpt, 2023. find
  38. L. Ouyang, J. Wu, X. Jiang, D. Almeida, C. Wainwright, P. Mishkin, C. Zhang, S. Agarwal, K. Slama, A. Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744, 2022. find
  39. B. Paranjape, S. Lundberg, S. Singh, H. Hajishirzi, L. Zettlemoyer, and M. T. Ribeiro. Art: Automatic multi-step reasoning and tool-use for large language models. arXiv preprint arXiv:2303.09014, 2023. find
  40. A. Parisi, Y. Zhao, and N. Fiedel. Talm: Tool augmented language models. arXiv preprint arXiv:2205.12255, 2022. find our card
  41. S. G. Patil, T. Zhang, X. Wang, and J. E. Gonzalez. Gorilla: Large language model connected with massive apis. arXiv preprint arXiv:2305.15334, 2023. find our card
  42. B. Peng, M. Galley, P. He, H. Cheng, Y. Xie, Y. Hu, Q. Huang, L. Liden, Z. Yu, W. Chen, et al. Check your facts and try again: Improving large language models with external knowledge and automated feedback. arXiv preprint arXiv:2302.12813, 2023. find our card
  43. J. Qian, H. Wang, Z. Li, S. Li, and X. Yan. Limitations of language models in arithmetic and symbolic induction. arXiv preprint arXiv:2208.05051, 2022. find
  44. Y. Qin, S. Hu, Y. Lin, W. Chen, N. Ding, G. Cui, Z. Zeng, Y. Huang, C. Xiao, C. Han, Y. R. Fung, Y. Su, H. Wang, C. Qian, R. Tian, K. Zhu, S. Liang, X. Shen, B. Xu, Z. Zhang, Y. Ye, B. Li, Z. Tang, J. Yi, Y. Zhu, Z. Dai, L. Yan, X. Cong, Y. Lu, W. Zhao, Y. Huang, J. Yan, X. Han, X. Sun, D. Li, J. Ph… find our card
  45. A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019. find
  46. S. Robertson, H. Zaragoza, et al. The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends® in Information Retrieval, 3(4):333–389, 2009. find our card
  47. T. L. Scao, A. Fan, C. Akiki, E. Pavlick, S. Ilić, D. Hesslow, R. Castagné, A. S. Luccioni, F. Yvon, M. Gallé, et al. Bloom: A 176b-parameter open-access multilingual language model. arXiv preprint arXiv:2211.05100, 2022. find
  48. T. Schick, J. Dwivedi-Yu, R. Dessì, R. Raileanu, M. Lomeli, L. Zettlemoyer, N. Cancedda, and T. Scialom. Toolformer: Language models can teach themselves to use tools. arXiv preprint arXiv:2302.04761, 2023. find our card
  49. Y. Shen, K. Song, X. Tan, D. Li, W. Lu, and Y. Zhuang. Hugginggpt: Solving ai tasks with chatgpt and its friends in huggingface. arXiv preprint arXiv:2303.17580, 2023. find our card
  50. W. Shi, S. Min, M. Yasunaga, M. Seo, R. James, M. Lewis, L. Zettlemoyer, and W.-t. Yih. Replug: Retrieval-augmented black-box language models. arXiv preprint arXiv:2301.12652, 2023. find our card
  51. N. Shinn, B. Labash, and A. Gopinath. Reflexion: an autonomous agent with dynamic memory and self-reflection. arXiv preprint arXiv:2303.11366, 2023. find our card
  52. H. Sun, Y. Zhuang, L. Kong, B. Dai, and C. Zhang. Adaplanner: Adaptive planning from feedback with language models, 2023. find
  53. J. Thorne, A. Vlachos, C. Christodoulopoulos, and A. Mittal. FEVER: a large-scale dataset for fact extraction and VERification. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), p… find our card
  54. H. Touvron, T. Lavril, G. Izacard, X. Martinet, M.-A. Lachaux, T. Lacroix, B. Rozière, N. Goyal, E. Hambro, F. Azhar, et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023. find
  55. X. Wang, S. Li, and H. Ji. Code4struct: Code generation for few-shot structured prediction from natural language. arXiv preprint arXiv:2210.12810, 2022. — [arXiv:2210.12810](https://arxiv.org/abs/2210.12810) find
  56. Z. Wang, S. Cai, A. Liu, X. Ma, and Y. Liang. Describe, explain, plan and select: Interactive planning with large language models enables open-world multi-task agents, 2023. find our card
  57. J. Wei, X. Wang, D. Schuurmans, M. Bosma, B. Ichter, F. Xia, E. Chi, Q. Le, and D. Zhou. Chain-of-thought prompting elicits reasoning in large language models. arXiv, page 2201.11903v6, 2022. find our card
  58. L. Weidinger, J. Mellor, M. Rauh, C. Griffin, J. Uesato, P.-S. Huang, M. Cheng, M. Glaese, B. Balle, A. Kasirzadeh, et al. Ethical and social risks of harm from language models. arXiv preprint arXiv:2112.04359, 2021. find our card
  59. S. Wiegreffe, J. Hessel, S. Swayamdipta, M. Riedl, and Y. Choi. Reframing human-ai collaboration for generating free-text explanations. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 632–658, 20… find
  60. S. Wolfram. Wolfram\|Alpha as the Way to Bring Computational Knowledge Superpowers to ChatGPT. Stephen Wolfram Writings, 2023. find
  61. C. Wu, S. Yin, W. Qi, X. Wang, Z. Tang, and N. Duan. Visual chatgpt: Talking, drawing and editing with visual foundation models. arXiv preprint arXiv:2303.04671, 2023. find our card
  62. Q. Xu, F. Hong, B. Li, C. Hu, Z. Chen, and J. Zhang. On the tool manipulation capability of open-source large language models. arXiv preprint arXiv:2305.16504, 2023. find
  63. R. Xu, Y. Yu, J. C. Ho, and C. Yang. Weakly-supervised scientific document classification via retrieval-augmented multi-stage training. arXiv preprint arXiv:2306.07193, 2023. — [arXiv:2306.07193](https://arxiv.org/abs/2306.07193) find
  64. Z. Yang, L. Li, J. Wang, K. Lin, E. Azarnasab, F. Ahmed, Z. Liu, C. Liu, M. Zeng, and L. Wang. Mm-react: Prompting chatgpt for multimodal reasoning and action. arXiv preprint arXiv:2303.11381, 2023. find
  65. Z. Yang, P. Qi, S. Zhang, Y. Bengio, W. Cohen, R. Salakhutdinov, and C. D. Manning. HotpotQA: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2369–2380, Brussels, Belgium, Oct.-Nov. 2018… find our card
  66. S. Yao, J. Zhao, D. Yu, N. Du, I. Shafran, K. R. Narasimhan, and Y. Cao. React: Synergizing reasoning and acting in language models. In The Eleventh International Conference on Learning Representations, 2023. find our card
  67. J. Zhang. Graph-toolformer: To empower llms with graph reasoning ability via prompt augmented by chatgpt, 2023. find
  68. M. Zhang and E. Choi. SituatedQA: Incorporating extra-linguistic contexts into QA. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7371–7387, Online and Punta Cana, Dominican Republic, Nov. 2021. Association for Computational Linguistics. find
  69. R. Zhang, Y. Yu, P. Shetty, L. Song, and C. Zhang. Prboost: Prompt-based rule discovery and boosting for interactive weakly-supervised learning. arXiv preprint arXiv:2203.09735, 2022. find
  70. Y. Zhuang, Y. Li, J. Zhang, Y. Yu, Y. Mou, X. Chen, L. Song, and C. Zhang. ReSel: Nary relation extraction from scientific text and tables by learning to retrieve and select. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 730–744, Abu Dhabi, United A… find