RiCoRecA: rich cooking recipe annotation schema

2026-01-12 · Source: Explosion · Developer tools and consulting for AI, Machine Learning and NLP - Explosion.ai · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Internet of Things (IoT) & Connected Devices, Data Science & Analytics · Depth: Expert, extended

Summary

RiCoRecA is a novel annotation schema designed to parse cooking recipes into a workflow representation suitable for automating tasks in "ambient kitchens" with smart devices. The schema involves information extraction tasks such as named entity recognition (NER), relation classification (RC), coreference resolution, and entity tracking, which are learned using a single joint model. Researchers created a dataset of 156 recipes annotated according to RiCoRecA and compared two transformer-based models, PEGASUS-X and LongT5, for parsing recipes into workflows. PEGASUS-X significantly outperformed LongT5 across all annotation tasks, achieving a 39% higher F-Score on average and demonstrating near human-like performance. This approach aims to enable IoT devices to interpret natural language recipes, create action plans, and collaborate seamlessly with users.

Key takeaway

For AI Scientists and Research Scientists developing smart kitchen systems, RiCoRecA offers a robust framework for transforming natural language recipes into executable IoT workflows. You should consider adopting this joint modeling approach with long transformers like PEGASUS-X to achieve high accuracy in named entity recognition, relation classification, and coreference resolution, which are critical for device automation. This methodology provides a pathway to overcome the limitations of isolated smart devices and realize truly ambient kitchen intelligence.

Key insights

RiCoRecA enables joint information extraction from recipes for IoT automation via a novel annotation schema and transformer models.

Principles

• Joint modeling improves information extraction.
• Long transformers handle extended text inputs.
• Workflows can automate IoT device actions.

Method

The method involves annotating recipes with NER, RC, and entity tracking using a custom Prodigy interface, then training encoder-decoder transformer models (PEGASUS-X, LongT5) on this dataset to parse raw text into semantically augmented workflows.

In practice

• Use PEGASUS-X for superior recipe workflow extraction.
• Annotate recipes with RiCoRecA for IoT automation.
• Leverage transformer models for structured prediction.

Topics

• RiCoRecA Schema
• Recipe Workflow Automation
• Natural Language Processing
• Transformer Models
• IoT Kitchens

Code references

• FilipposVentirozos/RicoReca
• FilipposVentirozos/RiCoRecA
• explosion/tokenizations

Best for: AI Scientist, Research Scientist, AI Researcher, NLP Engineer, Machine Learning Engineer

Related on AIssential

• A new version of OpenAI’s Codex is powered by a new dedicated chip — Dedicated AI hardware accelerates lightweight models for real-time developer collaboration and rapid iteration.
• Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings — Food ingredient embeddings can integrate both recipe co-occurrence and chemical compound data for richer representations.
• Chef Robotics Advances Bi-Manual Physical AI System for Prep Table Food Assembly Powered by a Food Foundation Model — Chef Robotics developed a bi-manual physical AI system for complex food assembly, powered by a Food Foundation Model.
• Kitchen Semantics — Build a kitchen vocabulary for AI assistants using simple, no-install tools.
• On GATE, Text and Social Media Analysis, and Detecting Misinformation Online — A multi-layer analytical pipeline effectively extracts and categorizes digital identity aspects of agrifood products from online discourse.
• Croissant Tasks: A Metadata Format for Reproducible Machine Learning Evaluations — Reproducibility in ML can be achieved via machine-actionable metadata, enabling agent-generated implementations.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Explosion · Developer tools and consulting for AI, Machine Learning and NLP - Explosion.ai.