Losses that Cook: Topological Optimal Transport for Structured Recipe Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918451353747456 |
|---|---|
| author | Ottoborgo, Mattia Cambrin, Daniele Rege Garza, Paolo |
| author_facet | Ottoborgo, Mattia Cambrin, Daniele Rege Garza, Paolo |
| contents | Cooking recipes are complex procedures that require not only a fluent and factual text, but also accurate timing, temperature, and procedural coherence, as well as the correct composition of ingredients. Standard training procedures are primarily based on cross-entropy and focus solely on fluency. Building on RECIPE-NLG, we investigate the use of several composite objectives and present a new topological loss that represents ingredient lists as point clouds in embedding space, minimizing the divergence between predicted and gold ingredients. Using both standard NLG metrics and recipe-specific metrics, we find that our loss significantly improves ingredient- and action-level metrics. Meanwhile, the Dice loss excels in time/temperature precision, and the mixed loss yields competitive trade-offs with synergistic gains in quantity and time. A human preference analysis supports our finding, showing our model is preferred in 62% of the cases. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_02531 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Losses that Cook: Topological Optimal Transport for Structured Recipe Generation Ottoborgo, Mattia Cambrin, Daniele Rege Garza, Paolo Computation and Language Artificial Intelligence Cooking recipes are complex procedures that require not only a fluent and factual text, but also accurate timing, temperature, and procedural coherence, as well as the correct composition of ingredients. Standard training procedures are primarily based on cross-entropy and focus solely on fluency. Building on RECIPE-NLG, we investigate the use of several composite objectives and present a new topological loss that represents ingredient lists as point clouds in embedding space, minimizing the divergence between predicted and gold ingredients. Using both standard NLG metrics and recipe-specific metrics, we find that our loss significantly improves ingredient- and action-level metrics. Meanwhile, the Dice loss excels in time/temperature precision, and the mixed loss yields competitive trade-offs with synergistic gains in quantity and time. A human preference analysis supports our finding, showing our model is preferred in 62% of the cases. |
| title | Losses that Cook: Topological Optimal Transport for Structured Recipe Generation |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2601.02531 |