CIIR@LiveRAG 2025: Optimizing Multi-Agent Retrieval Augmented Generation through Self-Training

Alireza Salemi, Mukta Maddipatla, Hamed Zamani
6/12/2025
cs.CLcs.IR

No official code found. Best implementation: muktac5/CIIR-LiveRAG (8 stars, MIT, last commit 2025-06-13).

Abstract

This paper presents mRAG, a multi-agent retrieval-augmented generation (RAG) framework composed of specialized agents for subtasks such as planning, searching, reasoning, and coordination. Our system uses a self-training paradigm with reward-guided trajectory sampling to optimize inter-agent collaboration and enhance response generation. Evaluated on DataMorgana-derived datasets during the SIGIR 2025 LiveRAG competition, mRAG outperforms conventional RAG baselines. We further analyze competition outcomes and showcase the framework's strengths with case studies, demonstrating its efficacy for complex, real-world RAG tasks.

View on arXivView PDF

Code Implementations(1)

muktac5/CIIR-LiveRAGCommunity100%
MIT

Codes for SIGIR'25 - LiveRAG competition

82PythonMay 12, 20251 years agoMIT

Similar Papers

RAG-Gym: Systematic Optimization of Language Agents for Retrieval-Augmented Generation

86% match2025

Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG

85% match1 repo1.7k2025

Improving Retrieval-Augmented Generation through Multi-Agent Reinforcement Learning

85% match1 repo1132025

SIRAG: Towards Stable and Interpretable RAG with A Process-Supervised Multi-Agent Framework

84% match2025

Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training

84% match2026

Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem

84% match2026

Cite this paper

@article{salemi2025ciirliverag,
  title  = {CIIR@LiveRAG 2025: Optimizing Multi-Agent Retrieval Augmented Generation through Self-Training},
  author = {Alireza Salemi and Mukta Maddipatla and Hamed Zamani},
  year   = {2025},
  eprint = {2506.10844},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2506.10844v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2506.10844.svg)](https://codeofpaper.com/paper/2506.10844)

Discussion