Exploring Large Language Models for Personalized Recipe Generation and Weight-Loss Management
Bibliographic record
Abstract
The emergence of large language models (LLMs) is transforming various health-related domains, including approaches to obesity management. Obesity remains one of the world’s leading health issues, prompting the research community to develop various weight-loss applications focused on physical activity, dietary planning, and related interventions. In this study, we explore the capability of the LLM ChatGPT for personalized dietary planning. We conducted two case studies: Case Study 1 examined self-supervised recipe generation using ChatGPT alone, while Case Study 2 investigated a self-supervised approach combining National Institute of Health standards with ChatGPT recipe recommendations. We also performed a user study to evaluate recipe recommendations from ChatGPT. Our results show that ChatGPT recommends appropriate recipes based on comparisons with the United States Department of Agriculture’s (USDA) recipe calculator. We found no significant difference between ChatGPT-generated recipe recommendation calories and USDA standards for either Case Study 1 (p = 0.8530) or Case Study 2 (p = 0.0687). In addition, we found significant weight loss in participants following these recipes in both Case Study 1 (p < 0.00001) and Case Study 2 (p = 0.0014). Furthermore, the user study with potential weight-loss participants revealed varying levels of satisfaction (p = 0.001) and identified themes related to meal preferences, effective prompt generation, and mixed concerns regarding privacy, trust, user consent, and data storage. We conclude by discussing additional findings from our case and user studies, and present opportunities, challenges, and design and ethical considerations for the research community.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.024 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".