Date
Publisher
arXiv
We present Aryabhata 1.0, a compact 7B parameter math reasoning model
optimized for the Indian academic exam, the Joint Entrance Examination (JEE).
Despite rapid progress in large language models (LLMs), current models often
remain unsuitable for educational use. Aryabhata 1.0 is built by merging strong
open-weight reasoning models, followed by supervised fine-tuning (SFT) with
curriculum learning on verified chain-of-thought (CoT) traces curated through
best-of-$n$ rejection sampling. To further boost performance, we apply
reinforcement learning with verifiable rewards (RLVR) using A2C objective with
group-relative advantage estimation along with novel exploration strategies
such as Adaptive Group Resizing and Temperature Scaling. Evaluated on both
in-distribution (JEE Main 2025) and out-of-distribution (MATH, GSM8K)
benchmarks, Aryabhata outperforms existing models in accuracy and efficiency,
while offering pedagogically useful step-by-step reasoning. We release
Aryabhata as a foundation model to advance exam-centric, open-source small
language models. This marks our first open release for community feedback
(https://huggingface.co/PhysicsWallahAI/Aryabhata-1.0); PW is actively training
future models to further improve learning outcomes for students.
Study at a glance
- Question: Can a compact AI model effectively solve Indian college entrance exam math problems with step-by-step reasoning?
- Setting and data: Math problems from the Indian Joint Entrance Examination (JEE Main 2025) and math test datasets (MATH, GSM8K).
- Method: Builds an AI model and evaluates its performance on standardized math exam questions.
Why it matters
Despite rapid progress in large language models, current models often remain unsuitable for educational use.
Key findings
- On India's Joint Entrance Examination in mathematics, the model outperformed existing models in both accuracy and efficiency.
- The model also outperformed existing models on broader mathematics tests beyond the targeted entrance exam.
- The tool generates pedagogically useful, step-by-step reasoning to explain mathematical solutions.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Who age?
Why use AI?
Study design
