NightFeats @ MMU-RAGent NeurIPS 2025: A Context-Optimized Multi-Agent RAG System for the Text-to-Text Track

Quentin Fever; Naziha Aslam

cs.CLApr 21, 2026

NightFeats @ MMU-RAGent NeurIPS 2025: A Context-Optimized Multi-Agent RAG System for the Text-to-Text Track

Authors: Quentin Fever, Naziha Aslam

Organizations: Team NightFeats

Abstract

We present NightFeats, a structured multi-agent retrieval-augmented generation (RAG) system submitted to the MMU-RAGent competition at NeurIPS 2025, where it was awarded Best Dynamic Evaluation in the text-to-text track. Rather than targeting benchmark maximization, this work proposes a principled pipeline that decomposes knowledge synthesis into three coordinated phases: retrieval, curation, and composition, each governed by explicit intermediate representations and handoff contracts. Inspired by Agentic Context Engineering (ACE), the system introduces temporal-semantic reranking, bounded contradiction reconciliation, and citation-preserving composition as core architectural primitives. Competition results show that NightFeats surpasses proprietary baselines including Claude-SonnetV2 and Nova-Pro on LLM-as-a-Judge and Human Likert evaluations, confirming that architectural transparency and verifiable evidence grounding are better aligned with human preferences than systems optimizing narrowly for automatic similarity metrics.

Explore similar work

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation
Apr 20, 2026Xingchen Xiao, Heyan Huang, Runheng Liu +1Retrieval-Augmented Generation Systems Hievi-Rag
Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
Jun 4, 2026Anuj Maharjan, Devinder Kaur, Richard MolyetRetrievers
Retrieval as Reasoning: Self-Evolving Agent-Native Retrieval via LLM-Wiki
May 25, 2026Haoliang Ming, Feifei Li, Xiaoqing Wu +1Agentic Retrieval Wikipedia

Apr 20, 2026 · cs.CLJ/K move · Enter open · S save

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation

Xingchen Xiao, Heyan Huang, Runheng Liu, Jincheng Xie

School of Computer Science and Technology, Beijing Institute of Technology · Department of Mathematical Sciences, Tsinghua University

Large language models (LLMs) are widely used in retrieval-augmented generation (RAG) to incorporate external knowledge at inference time. However, when retrieved contexts are noisy, incomplete, or heterogeneous, a single generation process often struggles to reconcile evidence effectively. We propose \textbf{MASS-RAG}, a multi-agent synthesis approach to retrieval-augmented generation that structures evidence processing into multiple role-specialized agents. MASS-RAG applies distinct agents for evidence summarization, evidence extraction, and reasoning over retrieved documents, and combines their outputs through a dedicated synthesis stage to produce the final answer. This design exposes multiple intermediate evidence views, allowing the model to compare and integrate complementary information before answer generation. Experiments on four benchmarks show that MASS-RAG consistently improves performance over strong RAG baselines, particularly in settings where relevant evidence is distributed across retrieved contexts.

Retrieval-Augmented Generation Systems Hievi-Rag Summarization

Open paper Ask about it PDF

Abstract

Explore similar work

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation

Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval

Retrieval as Reasoning: Self-Evolving Agent-Native Retrieval via LLM-Wiki