RecipeNet: A Hierarchical Transformer for Recipe Data
2026-08-14 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors studied how to better understand recipes, which are step-by-step instructions used in areas like making materials or medicines. They noticed that current methods simplify these recipes in ways that lose important details about how steps and parts in each step relate. To fix this, they created RecipeNet, a special model that looks closely at both the details within each step and how steps connect in order. Their tests showed RecipeNet works better than older methods at understanding recipes because it keeps track of this detailed structure.
Transformer architecturehierarchical modelingsequential dependenciestabular learningrecipe representationmaterials synthesispharmaceutical formulationstacked Transformer encodersstructured fieldsprocedural dependencies
Authors
Pin-Yen Huang, Sachin Chhabra, Prasanth Sai Gouripeddi, Abhinav Kumar, Baoxin Li
Abstract
Recipe data arises in domains such as materials synthesis, pharmaceutical formulation, and industrial manufacturing, where procedures are represented as ordered sequences of steps containing heterogeneous structured fields. Existing tabular learning methods typically flatten this structure into fixed-schema representations, limiting their ability to capture hierarchical field interactions and procedural dependencies. We propose RecipeNet, a hierarchical Transformer architecture that encodes field-level interactions within each step and sequential dependencies across steps through stacked Transformer encoders. Experiments on multiple recipe datasets and tasks demonstrate that RecipeNet consistently outperforms existing tabular models, highlighting the value of hierarchical and sequential modeling for recipe representation learning.