# Mitigating Misalignment Contagion by Steering with Implicit Traits

arXiv: https://arxiv.org/abs/2605.02751 · HTML: https://mlsanity.com/paper/2605.02751/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 4 May 2026.

## Builds on (most cited first, 9 of 9)

- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits](https://mlsanity.com/paper/2305.02547/) (2023)
- [InCharacter: Evaluating Personality Fidelity in Role-Playing Agents through Psychological Interviews](https://mlsanity.com/paper/2310.17976/) (2023)
- [Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs](https://mlsanity.com/paper/2502.17424/) (2025)
- [Improving Instruction-Following in Language Models through Activation Steering](https://mlsanity.com/paper/2410.12877/) (2024)
- [Subliminal Learning: Language models transmit behavioral traits via hidden signals in data](https://mlsanity.com/paper/2507.14805/) (2025)
- [Measuring and Controlling Instruction (In)Stability in Language Model Dialogs](https://mlsanity.com/paper/2402.10962/) (2024)
- [Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems](https://mlsanity.com/paper/2603.00131/) (2026)
- [Aligning Machiavellian Agents: Behavior Steering via Test-Time Policy Shaping](https://mlsanity.com/paper/2511.11551/) (2025)

## Built on by (most cited first, 1 of 1)

- [You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents](https://mlsanity.com/paper/2605.27586/) (2026)

## Abstract (arXiv)

Language models (LMs) are increasingly used in high-stakes, multi-agent settings, where following instructions and maintaining value alignment are critical. Most alignment research focuses on interactions between a single LM and a single user, failing to address the risk of misaligned behavior spreading between multiple LMs in multi-turn interactions. We find evidence of this phenomenon, which we call misalignment contagion, across multiple LMs as they engage multi-turn conversational social dilemma games. Specifically, we find that LMs become more anti-social after gameplay and that this effect is intensified when other players are steered to act maliciously. We explore different steering techniques to mitigate such misalignment contagion and find that reinforcing an LM's system prompt is insufficient and often harmful. Instead, we propose steering with implicit traits: a technique that intermittently injects system prompts with statements that reinforce an LMs initial traits and is more effective than system prompt repetition at keeping models in line with their initial pro-social behaviors. Importantly, this method does not require access to model parameters or internal model states, making it suitable for increasingly common use cases where complex multi-agent workflows are being designed with black box models.