policy

RLHF-Dialogue-Summarizer-with-PPO-and-Reward-Model

ivanluk914 · PyTorch

or hover any field below to flag it

Overview

Name
RLHF-Dialogue-Summarizer-with-PPO-and-Reward-Model
Author
ivanluk914
Framework
PyTorch
License
MIT
Skill type
other
Evidence level
untested
Task description
This project implements RLHF to fine-tune a FLAN-T5 language model for dialogue summarization. The training pipeline involves previous PEFT model, dataset preparation, reward modeling that leveraging a toxicity classifier, implementing PPO-based RL with trl library and evaluation before and after RL

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 RLHF-Dialogue-Summarizer-with-PPO-and-Reward-Model yet.

Datasets that reference this policy

0

No datasets reference RLHF-Dialogue-Summarizer-with-PPO-and-Reward-Model yet.