# Neural Program Synthesis with a Differentiable Fixer

arXiv: https://arxiv.org/abs/2006.10924 · HTML: https://mlsanity.com/paper/2006.10924/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Jun 2020.

## Built on by (most cited first, 6 of 6)

- [Latent Execution for Neural Program Synthesis](https://mlsanity.com/paper/2107.00101/) (2021)
- [Learning to Edit Visual Programs with Self-Supervision](https://mlsanity.com/paper/2406.02383/) (2024)
- [Gradient-Based Program Repair: Fixing Bugs in Continuous Program Spaces](https://mlsanity.com/paper/2505.17703/) (2025)
- [Neural Program Synthesis with Query](https://mlsanity.com/paper/2205.07857/) (2022)
- [Bootstrapping Fuzzers for Compilers of Low-Resource Language Dialects Using Language Models](https://mlsanity.com/paper/2512.05887/) (2025)
- [Latent Programming Horizons in Coding Agents](https://mlsanity.com/paper/2607.05188/) (2026)

## Abstract (arXiv)

We present a new program synthesis approach that combines an encoder-decoder based synthesis architecture with a differentiable program fixer. Our approach is inspired from the fact that human developers seldom get their program correct on the first attempt, and perform iterative testing-based program fixing to get to the desired program functionality. Similarly, our approach first learns a distribution over programs conditioned on an encoding of a set of input-output examples, and then iteratively performs fix operations using the differentiable fixer. The fixer takes as input the original examples and the current program's outputs on example inputs, and generates a new distribution over the programs with the goal of reducing the discrepancies between the current program outputs and the desired example outputs. We train our architecture end-to-end on the RobustFill domain, and show that the addition of the fixer module leads to a significant improvement on synthesis accuracy compared to using beam search.