Sweet Rl
Benchmark and research code for the paper SWEET-RL Training Multi-Turn LLM Agents onCollaborative Reasoning Tasks
About
SWEET-RL provides a framework for training large language model (LLM) agents to engage in multi-turn interactions for real-world tasks. It introduces a new benchmark, ColBench, which allows LLM agents to collaborate with human partners on backend programming and frontend design tasks. The implementation focuses on optimizing the training process through a novel reinforcement learning algorithm that enhances performance in collaborative content creation.
Open Source Health
- Stars
- 271
- Forks
- 12
- License
- Not stated
- Last commit
- 1 years ago
Resources & Links
Related Categories
Vendor
Facebookresearch
Publisher of Fairseq
Quick Links
Open Source
More by Facebookresearch
Related Products
RPG_KDD2025
This repository provides the code for implementing RPG described in our KDD'25 paper "Generating Long Semantic IDs in Parallel for Recommendation".
More from this vendor
Dagger
the missing software stack for CI
Top category match
Seamless Communication
Foundational Models for State-of-the-Art Speech and Text Translation
More from this vendor
OrienterNet
Source Code for Paper "OrienterNet Visual Localization in 2D Public Maps with Neural Matching"
More from this vendor
Matrix
Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data generation.
More from this vendor