shreyansh12183/shreyansh-hinglish-english-stem-500k

🇮🇳 Vigyan Indic-STEM: 500k Bilingual Hinglish & English Reasoning Corpus

Vigyan Indic-STEM 500k is a specialized, large-scale bilingual dataset created to bridge the pedagogical divide in STEM education across India. It pairs rigorous English first-principles scientific derivations with natural, conversational Hinglish (Hindi written in Roman script) explanations.

📖 Overview

In Tier-2 and Tier-3 educational institutions across India, STEM concepts (Physics, Chemistry, Calculus, Numerical Methods) are primarily taught using code-switched Hindi and English. Standard Western LLMs fail to communicate effectively in this natural register, either generating overly formal Devanagari or pure English.

This dataset provides 500,000 conversational STEM reasoning pairs designed to train and benchmark student-facing Small Language Models (SLMs).

Key Features

• —Bilingual Pedagogy: Formal scientific proofs in standard English paired with conversational step-by-step Hindi-English explanations.
• —Subjects Covered: Classical Mechanics, Thermodynamics, Organic & Inorganic Chemistry, Calculus, Differential Equations, and Semiconductor Fundamentals.
• —Curated by: Shreyansh Singh & Vigyan AI Research Labs.

📊 Dataset Structure

Each record contains:

• —instruction: The conceptual or numerical problem statement in English or Hinglish.
• —hought: First-principles reasoning chain decomposing the problem.
• — esponse: The bilingual pedagogical explanation terminating with the final verified solution.

📜 Citation

ibtex @dataset{singh2026vigyan_hinglish_500k, author = {Shreyansh Singh}, title = {Vigyan Indic-STEM: 500k Bilingual Hinglish Reasoning Corpus}, year = {2026}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/datasets/shreyansh12183/shreyansh-hinglish-english-stem-500k}} }