Newest papers

Small Average Error, Wide Individual Error: General-Purpose LLMs Estimating Energy Intake from Meal Photos in a Feeding Trial

Xiaomeng Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26538 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Perkiraan rata-rata interval makanan yang didasarkan pada titik-titik tidak dapat mempermudah penilaian makanan, dan model bahasa besar multi-purpose (LLM) harus membuat perkiraan sedemikian mudah diperoleh. Kebanyakan penilaian menggunakan satu foto yang diambil sebelum makan dan membandingkan dengan asupan energi yang disajikan, sementara penilaian diet hanya membutuhkan energi yang dimakan. Kami menilai enam model dari dua vendor (Claude Haiku, Sonnet dan Opus 5.5; GPT-6 luna, 6.1 sol dan 6 astra) pada 802 makanan rendah dari ACETADA, sebuah percobaan makan terkontrol dengan asupan energi yang dipertaruhkan, dengan foto sebelumnya hanya dipertimbangkan, sebelum dan setelah foto, dalam panggilan terpisah (subtraksi), atau kedua foto dalam satu permintaan. Tiga hasil berlari melawan intuisi. Pengurangan bukti kesalahan yang kurang dipertimbangkan

Machine translation (Meta M2M100); the English original follows.

Abstract

Photo-based estimation of food energy could ease dietary assessment, and general-purpose multimodal large language models (LLMs) make such estimates easy to obtain. Most evaluations use one photo taken before eating and compare with energy served, while dietary assessment needs energy eaten. We evaluated six models from two vendors (Claude Haiku, Sonnet and Opus 5.5; GPT-6 luna, 6.1 sol and 6 astra) on 802 meals from ACETADA, a controlled feeding trial with weighed intake, with the before photo only, before and after photos in separate calls (subtraction), or both photos in one request. Three results ran against intuition. Subtraction under-estimated intake for five of six models, because the models still report hundreds of kilocalories for a nearly empty tray. Scaling the before-only estimate by the average share eaten (0.87) matched much of the gain from the after photo; beyond that constant, the second photo lowered meal-level error by only 2.2 to 3.4 points, mainly in meals with food left. With one photo, two models had a near-zero intake bias only because they under-estimated the served energy and ignored uneaten food. In ablations, adding the after photo improved accuracy under both prompt formats for five models, while prompt effects varied across models. Over 148 feeding days, three models kept the 90% confidence interval of their mean error within a pre-set 10% margin, but no model reached the pre-set 30% limits of agreement. Protein was under-estimated by every model, most in protein-dense meals; repeated calls changed error by at most 1.4 points; a high reasoning effort helped only the smallest model (4.2 points) and changed the others by less than 1 point. A low average bias is not evidence of accurate individual intake estimation; such estimates should support, not replace, assessment by a dietitian.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Calorie, Meal, Statistics, Energy (signal processing), Food intake, Food energy

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange