Saved in:
Bibliographic Details
Main Authors: Wu, Yuanhong, Bouneffouf, Djallel, Hsu, D. Frank
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.11126
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911506791137280
author Wu, Yuanhong
Bouneffouf, Djallel
Hsu, D. Frank
author_facet Wu, Yuanhong
Bouneffouf, Djallel
Hsu, D. Frank
contents Aligning large language models (LLMs) with human values is a central challenge for ensuring trustworthy and safe deployment. While existing methods such as Reinforcement Learning from Human Feedback (RLHF) and its variants have improved alignment, they often rely on a single evaluator or narrowly defined reward signals, limiting their ability to capture ethical pluralism. In this work, we propose the Value Alignment System using Combinatorial Fusion Analysis (VAS-CFA), a framework that operationalizes multi-agent fusion alignment. It instantiates multiple moral agents, each fine-tuned to represent a distinct normative perspective, and fuses their outputs using CFA with both rank- and score-based aggregation. This design leverages cognitive diversity, between agents, to mitigate conflicts and redundancies across multiple agents, producing responses that better reflect human values. Empirical evaluation demonstrates that VAS-CFA outperforms both single agent baselines and prior aggregation approaches on standard metrics, showing that multi-agent fusion provides a robust and effective mechanism for advancing value alignment in LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2603_11126
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Enhancing Value Alignment of LLMs with Multi-agent system and Combinatorial Fusion
Wu, Yuanhong
Bouneffouf, Djallel
Hsu, D. Frank
Multiagent Systems
Computation and Language
Aligning large language models (LLMs) with human values is a central challenge for ensuring trustworthy and safe deployment. While existing methods such as Reinforcement Learning from Human Feedback (RLHF) and its variants have improved alignment, they often rely on a single evaluator or narrowly defined reward signals, limiting their ability to capture ethical pluralism. In this work, we propose the Value Alignment System using Combinatorial Fusion Analysis (VAS-CFA), a framework that operationalizes multi-agent fusion alignment. It instantiates multiple moral agents, each fine-tuned to represent a distinct normative perspective, and fuses their outputs using CFA with both rank- and score-based aggregation. This design leverages cognitive diversity, between agents, to mitigate conflicts and redundancies across multiple agents, producing responses that better reflect human values. Empirical evaluation demonstrates that VAS-CFA outperforms both single agent baselines and prior aggregation approaches on standard metrics, showing that multi-agent fusion provides a robust and effective mechanism for advancing value alignment in LLMs.
title Enhancing Value Alignment of LLMs with Multi-agent system and Combinatorial Fusion
topic Multiagent Systems
Computation and Language
url https://arxiv.org/abs/2603.11126