---
title: "Surya Institute's 86M Model Beats Giants 100x Its Size at Arithmetic"
subtitle: "An 86M-parameter model trained on an Indonesian teaching method beats models 100x its size at basic arithmetic, revealing how structured pedagogy shapes neural computation"
url: "https://alphasignal.ai/news/surya-institute-s-86m-model-beats-giants-100x-its-size-at-arithmetic"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-06-03T00:00:00.000Z"
updated: "2026-06-11T02:59:26.088Z"
type: "Paper"
topics:
  - "Post Training"
  - "Reasoning"
subtopics:
  - "Fine Tuning"
  - "Math Reasoning"
upvotes: 1360
source_url: "https://arxiv.org/abs/2606.05106"
comments: 0
truncated: true
---

# Surya Institute's 86M Model Beats Giants 100x Its Size at Arithmetic

> An 86M-parameter model trained on an Indonesian teaching method beats models 100x its size at basic arithmetic, revealing how structured pedagogy shapes neural computation

What if the secret to teaching a language model arithmetic was to borrow from a classroom in Indonesia? That is the central bet of a new [paper from researchers at the Surya Institute](https://arxiv.org/abs/2606.05106), who trained a tiny 86-million-parameter GPT-2-style model to do arithmetic using a human teaching method called GASING -- and watched it outperform models more than 100 times its size.

The result is not just a performance number. The paper offers one of the clearest windows yet into **how** a language model actually learns to compute, tracing the emergence of arithmetic skill through three distinct phases and linking each phase to measurable changes inside the network.

## Why arithmetic is still a hard problem

Language models have a well-known arithmetic problem. LLMs can be surprisingly bad at arithmetic, often due to the arbitrary way digits are tokenized. Standard tokenizers, built to compress English text efficiently, treat numbers inconsistently: the string "128" might become one token, two tokens, or three, depending on context. That inconsistency makes it hard for a model to learn positional relationships between digits, which are essential for carrying, borrowing, and place-value reasoning.

The dominant fix has been scale. Chain-of-thought prompting is an emergent ability of scale -- it requires a sufficiently large language model to improve performance, and actually hurts performance for small language models. That creates a frustrating situation: the models that most need structured reasoning help are the ones that benefit least from it.

## The GASING method: a classroom technique repurposed for tokens

GASING stands for Gampang, Asyik, Menyenangkan -- Easy, Fun, Enjoyable -- and covers all four basic arithmetic operations: addition, multiplication, subtraction, and division. It was discovered by Professor Yohanes Surya and has been widely used across Indonesia to address numeracy gaps in under-resourced communities.

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [Arithmetic Pedagogy for Language Models (arXiv 2606.05106)](https://arxiv.org/abs/2606.05106)
- [Arithmetic Pedagogy for Language Models -- Full HTML](https://arxiv.org/html/2606.05106)
- [Syllabic Agglutinative Tokenizations for Indonesian LLM (arXiv 2601.11643)](https://arxiv.org/html/2601.11643)
- [Adaptive Engram Memory System for Indonesian Language Model: TOBA-LM (arXiv 2603.10006)](https://arxiv.org/abs/2603.10006)
- [The Matematika GASING Method as an Intervention for Improving Arithmetic Fluency](https://ejournal.uncen.ac.id/index.php/JIPI/article/view/5368)
