# ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systems

arXiv: https://arxiv.org/abs/2602.08567 · HTML: https://mlsanity.com/paper/2602.08567/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 16 of its 34 references that are in the map.
- First posted on arXiv 9 Feb 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 16 of 16)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challenges](https://mlsanity.com/paper/2401.07339/) (2024)
- [Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration](https://mlsanity.com/paper/2307.05300/) (2023)
- [A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems](https://mlsanity.com/paper/2402.18013/) (2024)
- [Simulating Opinion Dynamics with Networks of LLM-based Agents](https://mlsanity.com/paper/2311.09618/) (2023)
- [Multi-Agent Design: Optimizing Agents with Better Prompts and Topologies](https://mlsanity.com/paper/2502.02533/) (2025)
- [Who is GPT-3? An Exploration of Personality, Values and Demographics](https://mlsanity.com/paper/2209.14338/) (2022)
- [Improving Multi-Agent Debate with Sparse Communication Topology](https://mlsanity.com/paper/2406.11776/) (2024)
- [Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimization](https://mlsanity.com/paper/2404.02183/) (2024)
- [Interpretable Deep Learning under Fire](https://mlsanity.com/paper/1812.00891/) (2018)
- [Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?](https://mlsanity.com/paper/2501.15463/) (2025)
- [Can Language Models Reason about Individualistic Human Values and Preferences?](https://mlsanity.com/paper/2410.03868/) (2024)
- [Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing](https://mlsanity.com/paper/2406.14230/) (2024)
- [Improving fairness in speaker verification via Group-adapted Fusion Network](https://mlsanity.com/paper/2202.11323/) (2022)

## Built on by (most cited first, 4 of 4)

- [Alignment as Iatrogenesis: Pastoral Power, Collective Pathology, and the Structural Limits of Monolingual Safety Evaluation](https://mlsanity.com/paper/2603.08723/) (2026)
- [QUIVER: A Formal Framework for Quantifying Perturbation Propagation and Bifurcation in Compound AI Systems](https://mlsanity.com/paper/2605.23956/) (2026)
- [Pseudo-Deliberation in Language Models: When Reasoning Fails to Align Values and Actions](https://mlsanity.com/paper/2605.09893/) (2026)
- [The Failure Happens Before the Drift: The Social Dynamics of Values in LLM Agent Societies](https://mlsanity.com/paper/2609.05514/) (2026)

## Abstract (arXiv)

Multi-agent large language model (LLM) systems increasingly consist of agents that observe and respond to one another's outputs. While value alignment is typically evaluated for isolated models, how value perturbations propagate through agent interactions remains poorly understood. We present ValueFlow, a perturbation-based framework that measures value drift in multi-agent systems via a 56-value valuation dataset derived from the Schwartz Value Survey, with agent value orientations scored using an LLM-as-a-judge protocol. ValueFlow decomposes value drift into agent-level response behavior and system-level structural effects, captured by two metrics: beta-susceptibility, an agent's sensitivity to perturbed peer value signals, and system susceptibility (SS), the effect of node-level perturbations on final system outputs.Experiments span across value dimensions, backbones, personas, and topologies, showing that susceptibility varies sharply across values and is strongly shaped by interaction structure, indicating that value alignment in multi-agent systems is a system-level property, not just an agent-level one. ValueFlow thus provides a principled basis for auditing and mitigating value propagation in deployed multi-agent systems.