EMNLP Findings2024
Ask the experts: sourcing a high-quality nutrition counseling dataset through Human-AI collaboration
Simone Balloccu, Ehud Reiter, Karen Li, Rafael Sargsyan, Vivek Kumar, Diego Reforgiato Recupero, Daniele Riboni, Ondrej Dusek
Abstract
Large Language Models (LLMs) are being employed by end-users for various tasks, including sensitive ones such as health counseling, disregarding potential safety concerns. It is thus necessary to understand how adequately LLMs perform in such domains. We conduct a case study on ChatGPT in nutrition counseling, a popular use-case where the model supports a user with their dietary struggles. We crowdsource real-world diet-related struggles, then work with nutrition experts to generate supportive text using ChatGPT. Finally, experts evaluate the safety and text quality of ChatGPT's output. The result is the HAI-Coaching dataset, containing ∼2.4K crowdsourced dietary struggles and ∼97K corresponding ChatGPTgenerated and expert-annotated supportive texts. We analyse ChatGPT's performance, discovering potentially harmful behaviours, especially for sensitive topics like mental health. Finally, we use HAI-Coaching to test open LLMs on various downstream tasks, showing that even the latest models struggle to achieve good performance. HAI-Coaching
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text Data cites Language Models are Few-Shot Learners
- Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text Data cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity cites Language Models are Few-Shot Learners
Builds on13
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order SensitivityYao Lu, Max Bartolo, Alastair Moore, Sebastian Riedel et al.ACL 2022 · 1,494 citations
- Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text DataXuhai Xu, Bingsheng Yao, Yuanzhe Dong, Saadia Gabriel et al.UbiComp 2024 · 281 citations
- Evaluating Large Language Models in Generating Synthetic HCI Research Data: a Case StudyPerttu Hämäläinen, Mikke Tavast, Anton KunnariCHI 2023 · 244 citations
Related papers
- KokoroChat: A Japanese Psychological Counseling Dialogue Dataset Collected via Role-Playing by Trained CounselorsZhiyang Qi, Takumasa Kaneko, Keiko Takamizo, Mariko Ukiyo et al.ACL 2025 · 7 citations
- CounselBench: A Large-Scale Expert Evaluation and Adversarial Benchmarking of Large Language Models in Mental Health Question AnsweringYahan Li, Jifan Yao, John Bosco S. Bunyi, Adam C. Frank et al.ICLR 2026 · 24 citations
- 'I've talked to ChatGPT about my issues last night.': Examining Mental Health Conversations with Large Language Models through Reddit AnalysisKyuha Jung, Gyuho Lee, Yuanhui Huang, Yunan ChenCSCW 2025 · 23 citations
- GPTCoach: Towards LLM-Based Physical Activity CoachingMatthew Jörke, Shardul Sapkota, Lyndsea Warkenthien, Niklas Vainio et al.CHI 2025 · 89 citations
- Towards Interpretable Mental Health Analysis with Large Language ModelsKailai Yang, Shaoxiong Ji, Tianlin Zhang, Qianqian Xie et al.EMNLP 2023 · 114 citations
