LLM reasoning traces can be compiled into reusable symbolic solvers that achieve high accuracy on program synthesis benchmarks at zero inference cost and transfer to other domains.
Deepcoder: Learning to write programs.arXiv preprint arXiv:1611.01989
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
We develop a first line of attack for solving programming competition-style problems from input-output examples using deep learning. The approach is to train a neural network to predict properties of the program that generated the outputs from the inputs. We use the neural network's predictions to augment search techniques from the programming languages community, including enumerative search and an SMT-based solver. Empirically, we show that our approach leads to an order of magnitude speedup over the strong non-augmented baselines and a Recurrent Neural Network approach, and that we are able to solve problems of difficulty comparable to the simplest problems on programming competition websites.
years
2026 3verdicts
UNVERDICTED 3representative citing papers
CP-Agent improves LLM competitive programming performance via calibrated feedback mechanisms that target false-admission risk, evidence against bad programs, and success hazard.
Children and LLM agents show parallel adaptations to evidence reliability in a Bayesian program induction task but differ in information-seeking costs and compliance.
citing papers explorer
-
ReaComp: Compiling LLM Reasoning into Symbolic Solvers for Efficient Program Synthesis
LLM reasoning traces can be compiled into reusable symbolic solvers that achieve high accuracy on program synthesis benchmarks at zero inference cost and transfer to other domains.
-
CP-Agent: A Calibrated Risk-Controlled Agent for Feedback-Driven Competitive Programming
CP-Agent improves LLM competitive programming performance via calibrated feedback mechanisms that target false-admission risk, evidence against bad programs, and success hazard.
-
Hypothesis Generation and Inductive Inference in Children and Language Models
Children and LLM agents show parallel adaptations to evidence reliability in a Bayesian program induction task but differ in information-seeking costs and compliance.