Datasetpublic

mistral-hackaton-2026/robuchan-data

Robuchan Dataset Synthetic dietary recipe adaptation dataset for fine-tuning language models. Each example is a chat-format conversation where a user provides a recipe and dietary restriction, and the assistant produces a structured adaptation. Generated for the Mistral AI Worldwide Hackathon Tokyo (Feb 28 - Mar 1, 2026). Associated model: sumitdotml/robuchan Dataset Structure Splits Split Rows Purpose train 1,090 Fine-tuning training set… See the full description on the dataset page: https://huggingface.co/datasets/mistral-hackaton-2026/robuchan-data.

sourceHugging Faceapache-2.0updated 7mo agoView on Hugging Face

0likes41downloads

Dataset Card

Robuchan Dataset

Synthetic dietary recipe adaptation dataset for fine-tuning language models. Each example is a chat-format conversation where a user provides a recipe and dietary restriction, and the assistant produces a structured adaptation.

Generated for the Mistral AI Worldwide Hackathon Tokyo (Feb 28 - Mar 1, 2026).

Associated model: `sumitdotml/robuchan`

Dataset Structure

Splits

Split	Rows	Purpose
`train`	1,090	Fine-tuning training set
`validation`	122	Fine-tuning validation set
`eval_quick50`	50	Quick evaluation gate
`eval_final150`	150	Full evaluation freeze
`eval_hard_cases`	30	Curated difficult adaptations

Schema

Each row is a Mistral chat-format object with a messages array containing three roles:

—system: Sets the assistant's priorities (dietary compliance > dish identity > practicality) and defines the required output sections.
—user: Provides the recipe (title, ingredients with quantities, steps) and the target dietary restriction.
—assistant: Returns a structured adaptation with 5 sections.

Output Sections

Section	Content
Substitution Plan	One row per banned ingredient: `original -> replacement (rationale)`
Adapted Ingredients	Full ingredient list with quantities — no placeholders
Adapted Steps	Complete numbered cooking steps reflecting all substitutions
Flavor Preservation Notes	3+ notes on how taste/texture/aroma are maintained
Constraint Check	Explicit checklist confirming all violations resolved

Dietary Restrictions (train split)

Restriction	Rows
vegetarian	791
vegan	182
dairy-free	76
other	41

The eval splits additionally cover gluten-free, low-sodium, low-sugar, nut-free, egg-free, shellfish-free, and low-fat.

Generation Pipeline

1.Source pool: 530K recipes from Food.com, filtered for parseable ingredients/steps and at least one dietary violation.
2.Synthetic generation: Each source recipe is paired with a dietary constraint and sent to mistral-large-latest to produce an adapted version.
3.Quality gate: Deterministic checks reject candidates that fail any of:
4.Constraint compliance (all banned ingredients removed)
5.Structural completeness (all 5 sections present, no ... placeholders)
6.Ingredient parseability (quantities and units present)
7.Violation coverage (every detected violation mapped in Substitution Plan)
8.Single-candidate policy: One generation attempt per recipe; drop on fail.

Prompt Templates

Three user prompt templates with identical semantics prevent format overfitting:

Template	Share	Style
A — Labeled Block	50%	Structured labeled fields
B — Natural Request	30%	Conversational prose
C — Goal-Oriented	20%	Goal-first with bullet lists

Template assignment is deterministic: hash(source_recipe_id + restriction) % 100.

Supporting Files

File	Description
`eval/constraints.json`	Banned ingredient lists per dietary category
`eval/category_aliases.json`	Category name normalization
`kb/swaps_v0.json`	Curated ingredient swap rules (20+ rules)

Usage

python

from datasets import load_dataset

ds = load_dataset("mistral-hackaton-2026/robuchan-data")
print(ds["train"][0]["messages"])

For fine-tuning with Mistral API, use the JSONL files directly:

bash

# Each line is {"messages": [{"role": "system", ...}, {"role": "user", ...}, {"role": "assistant", ...}]}
head -1 data/train_filtered.jsonl | python -m json.tool

Authors

—sumitdotml
—Kaustubh Hiware

Citation

bibtex

@misc{robuchan2026,
  title  = {Robuchan: Recipe Dietary Adaptation via Fine-Tuned Ministral-8B},
  author = {sumitdotml and Hiware, Kaustubh},
  year   = {2026},
  url    = {https://huggingface.co/datasets/mistral-hackaton-2026/robuchan-data}
}