2000 character limit reached
Successfully Guiding Humans with Imperfect Instructions by Highlighting Potential Errors and Suggesting Corrections
Published 26 Feb 2024 in cs.AI, cs.CL, and cs.HC | (2402.16973v2)
Abstract: LLMs will inevitably err in situations with which they are unfamiliar. However, by effectively communicating uncertainties, they can still guide humans toward making sound decisions in those contexts. We demonstrate this idea by developing HEAR, a system that can successfully guide humans in simulated residential environments despite generating potentially inaccurate instructions. Diverging from systems that provide users with only the instructions they generate, HEAR warns users of potential errors in its instructions and suggests corrections. This rich uncertainty information effectively prevents misguidance and reduces the search space for users. Evaluation with 80 users shows that HEAR achieves a 13% increase in success rate and a 29% reduction in final location error distance compared to only presenting instructions to users. Interestingly, we find that offering users possibilities to explore, HEAR motivates them to make more attempts at the task, ultimately leading to a higher success rate. To our best knowledge, this work is the first to show the practical benefits of uncertainty communication in a long-horizon sequential decision-making problem.
- The hcrc map task corpus. Language and speech, 34(4):351–366.
- Vision-and-language navigation: Interpreting visually-grounded navigation instructions in real environments. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3674–3683.
- Does the whole exceed its parts? the effect of ai explanations on complementary team performance. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, pages 1–16.
- Nina L Corvelo Benz and Manuel Gomez Rodriguez. 2023. Human-aligned calibration for ai-assisted decision making. arXiv preprint arXiv:2306.00074.
- To trust or to think: cognitive forcing functions can reduce overreliance on ai in ai-assisted decision-making. Proceedings of the ACM on Human-Computer Interaction, 5(CSCW1):1–21.
- Report on the First NLG Challenge on Generating Instructions in Virtual Environments (GIVE). In Proceedings of the 12th European Workshop on Natural Language Generation (ENLG 2009), pages 165–173, Athens, Greece. Association for Computational Linguistics.
- Improving faithfulness in abstractive summarization with contrast candidate generation and selection. arXiv preprint arXiv:2104.09061.
- Detecting and mitigating hallucinations in machine translation: Model internal workings alone do well, sentence similarity even better. arXiv preprint arXiv:2212.08597.
- Feqa: A question answering evaluation framework for faithfulness assessment in abstractive summarization. arXiv preprint arXiv:2005.03754.
- Ranking generated summaries by correctness: An interesting but challenging application for natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 2214–2220.
- Unified pragmatic models for generating and following instructions. arXiv preprint arXiv:1711.04987.
- Speaker-follower models for vision-and-language navigation. Advances in Neural Information Processing Systems, 31.
- Robert Goeddel and Edwin Olson. 2012. Dart: A particle-based method for generating easy-to-follow directions. In 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 1213–1219. IEEE.
- Airbert: In-domain pretraining for vision-and-language navigation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1634–1643.
- Jessica Hullman. 2019. Why authors don’t visualize uncertainty. IEEE transactions on visualization and computer graphics, 26(1):130–139.
- Adam Tauman Kalai and Santosh S Vempala. 2023. Calibrated language models must hallucinate. arXiv preprint arXiv:2311.14648.
- A new path: Scaling vision-and-language navigation with synthetic instructions and imitation learning. arXiv preprint arXiv:2210.03112.
- Report on the second nlg challenge on generating instructions in virtual environments (give-2). In Proceedings of the 6th international natural language generation conference. The Association for Computer Linguistics.
- Evaluating the factual consistency of abstractive text summarization. arXiv preprint arXiv:1910.12840.
- Towards a science of human-ai decision making: a survey of empirical studies. arXiv preprint arXiv:2112.11471.
- A token-level reference-free hallucination detection benchmark for free-form text generation. arXiv preprint arXiv:2104.08704.
- Who should i trust: Ai or myself? leveraging human and ai correctness likelihood to promote appropriate trust in ai-assisted decision-making. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, pages 1–19.
- Improving vision-and-language navigation with image-text pairs from the web. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VI 16, pages 259–274. Springer.
- Identifying fluently inadequate output in neural and statistical machine translation. In Proceedings of Machine Translation Summit XVII: Research Track, pages 233–243.
- On faithfulness and factuality in abstractive summarization. arXiv preprint arXiv:2005.00661.
- Domain robustness in neural machine translation. arXiv preprint arXiv:1911.03109.
- Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1):5485–5551.
- Deciding fast and slow: The role of cognitive biases in ai-assisted decision-making. Proceedings of the ACM on Human-Computer Interaction, 6(CSCW1):1–22.
- Controlling hallucinations at word level in data-to-text generation. Data Mining and Knowledge Discovery, pages 1–37.
- Object hallucination in image captioning. arXiv preprint arXiv:1809.02156.
- Cheap talk and credibility: The consequences of confidence and accuracy on advisor credibility and persuasiveness. Organizational Behavior and Human Decision Processes, 121(2):246–255.
- Report on the second second challenge on generating instructions in virtual environments (give-2.5). In 13th European workshop on natural language generation.
- Neural machine translation incorporating named entity. In Proceedings of the 27th International Conference on Computational Linguistics, pages 3240–3250.
- Uncalibrated models can improve human-ai collaboration. Advances in Neural Information Processing Systems, 35:4004–4016.
- Chaojun Wang and Rico Sennrich. 2020. On exposure bias, hallucination and domain shift in neural machine translation. arXiv preprint arXiv:2005.03642.
- Less is more: Generating grounded navigation instructions from landmarks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 15428–15438.
- Neural text generation with unlikelihood training. arXiv preprint arXiv:1908.04319.
- Challenges in data-to-document generation. arXiv preprint arXiv:1707.08052.
- Understanding and detecting hallucinations in neural machine translation via model introspection. arXiv preprint arXiv:2301.07779.
- Effect of confidence and explanation on accuracy and trust calibration in ai-assisted decision making. In Proceedings of the 2020 conference on fairness, accountability, and transparency, pages 295–305.
- Define, evaluate, and improve task-oriented cognitive capabilities for instruction generation models. In Findings of ACL.
- Hallucination detection for grounded instruction generation. In Findings of the Empirical Methods in Natural Language Processing: EMNLP 2023, Singapore. Association for Computational Linguistics.
- Detecting hallucinated content in conditional neural sequence generation. arXiv preprint arXiv:2011.02593.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.