MathFormer: Testing whether symbolic math is pattern matching or reasoning [D]

Excerpt

Repo link and results - https://github.com/Abhinand20/MathFormer Task: Given a factorized expression like (7-3*z)*(-5*z-9), predict the expanded form -> 15*z\*2-8\*z-63 Key takeaway: A tiny (4M param) seq2seq model trained with no math knowledge reaches ~98.6% accuracy on symbolic math tasks, suggesting it learns structural token transformations rather than any notion of operators or variabl