SyncAI.news, a Varaisys broadcasting
Energy estimates of open-weight large language models across stated food weights: a dose-response evaluation
DR

Donato Riccio, Fabrizio Maturo, Iulia Brezeanu

· 1 min read

ResearcharXiv cs.CL

Energy estimates of open-weight large language models across stated food weights: a dose-response evaluation

arXiv:2611.25976v1 Announce Type: new Abstract: We tested whether energy estimates of open-weight large language models (LLMs) scale proportionally with an exactly stated food weight, and whether computing energy density times weight removes the error. Under a prespecified protocol, seven open-weight LLMs estimated the energy of 168 randomly drawn foods from the USDA Food and Nutrient Database for Dietary Studies at nine stated weights, from 0.25 to 4 times the default amount. The primary outcome was the elasticity of the estimate with respect to weight (1: proportional scaling). A two-step estimate multiplied each model's own stated energy density by the weight in code. All seven models compressed their estimates, with elasticities from 0.548 to 0.979 (every 95% CI below 1); four compressed severely (0.548 to 0.709). These four overestimated a quarter of the default amount by 47% to 130% and underestimated four times it by 20% to 34%. Supplying the density restored near-proportional scaling in three of the four, and the two-step estimate cut the typical error factor at these extremes from 1.47-2.43 to 1.20-1.38. Compression persisted at full precision and grew in five of seven models under a plain question. The models hold the energy densities needed for proportional scaling and answer near a typical serving, reproducing the flat-slope syndrome of dietary recall. Accuracy at usual portions does not reveal it. Food-logging tools should compute the energy of weighed entries as density times weight, and evaluations should test fixed foods across stated weights.

Original source

This story was published by arXiv cs.CL and written by Donato Riccio, Fabrizio Maturo, Iulia Brezeanu. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News