… Note that our execution-guided synthesis algorithm can be applied to any neural synthesizer
Γ, … In our evaluation, we demonstrate that our execution-guided synthesis technique helps …

… line of work on execution-guided program synthesis [47, 18, … We also compare with existing
neural program synthesis models with … RobustFill [17] is the state-of-the-art neural network …

… The second synthesizer is the execution-guided neural program synthesis model
proposed in [3], denoted as EGNPS. The model architecture of EGNPS similar to LGRL, but it …

… to tackle the problem of program synthesis is an old idea … executionguided neural program
synthesis,’ independently proposed by [12] and [13], where a neural network writes a program …

… neural program synthesis that performs decomposition within the execution space. A PBE task
defines a program by pairs of program … Execution-guided synthesis is a popular form of this…

… We propose a new approach, model predictive program synthesis (MPPS), that uses program
synthesis to automatically generate the guiding programs. It trains a generative model to …

… to ours is neural execution-guided synthesis, … programs considered as well as wall-clock
time. In fact, showing that learning-in-the-loop can be made fast enough for program synthesis …

… This paper introduces EVAPS, a novel approach to augment the applicability of neural
program synthesis by integrating partial environmental observations. By utilizing both the …

… (2021) note that large language models are not good at modeling execution semantics of
programs; we see our execution-guided pruning techniques as a path forward in this domain. …

… challenging program synthesis tasks on list manipulation. The experiments show promising
… In this section, we formally state our target problem of neural-guided program synthesis from …