Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
$\mathcal{B}$-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis
2024-01-16
· ICLR 2024 spotlight ·
anchor
Findings
IC-1231
On APPS, released code generation models span pass@1 from 0.20 (GPT-3 175B) to 6.20 (CodeRL), with value-based and policy-based RL methods outperforming supervised baselines