policy

SLMFIX-Leveraging-Small-LLMs-for-Error-Fixing-with-Reinforcement-Learning

nihaal237 · PyTorch

or hover any field below to flag it

Overview

Name
SLMFIX-Leveraging-Small-LLMs-for-Error-Fixing-with-Reinforcement-Learning
Author
nihaal237
Framework
PyTorch
License
unknown
Skill type
manipulation
Evidence level
untested
Task description
SLMFix — Uses PPO-based RL to train Qwen-2.5-Coder-0.5B to fix broken SQL and Bash commands generated by a larger proposer model. Broken candidates are detected via SQLGlot and bashlex validators, rewarded using AST and token-level F1 similarity, and evaluated on Spider and NL2Bash benchmarks.

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

3+17 mentioned but not in catalog yet

Compatible environments

0

No environments list SLMFIX-Leveraging-Small-LLMs-for-Error-Fixing-with-Reinforcement-Learning yet.

Datasets that reference this policy

0

No datasets reference SLMFIX-Leveraging-Small-LLMs-for-Error-Fixing-with-Reinforcement-Learning yet.