---
title: "RoleCDE Catches Every Major LLM Breaking Character Under Moral Pressure"
subtitle: "RoleCDE exposes a systematic flaw in LLM role-playing agents: they default to their alignment training instead of staying in character when values conflict."
url: "https://alphasignal.ai/news/rolecde-catches-every-major-llm-breaking-character-under-moral-pressure"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-06-01T00:00:00.000Z"
updated: "2026-06-09T08:11:33.237Z"
type: "Paper"
topics:
  - "Benchmarks"
  - "Post Training"
subtopics:
  - "Alignment"
  - "Fine Tuning"
upvotes: 2381
source_url: "https://arxiv.org/abs/2606.01552"
comments: 0
truncated: true
---

# RoleCDE Catches Every Major LLM Breaking Character Under Moral Pressure

> RoleCDE exposes a systematic flaw in LLM role-playing agents: they default to their alignment training instead of staying in character when values conflict.

Ask an LLM to play a ruthless corporate negotiator, a morally grey detective, or a historical figure with views that clash with modern ethics, and something predictable happens: the model blinks. Instead of committing to the role, it quietly reverts to its safety-trained, morality-consistent self. A new benchmark called [RoleCDE](https://arxiv.org/abs/2606.01552) puts a name to this behavior, measures it at scale, and offers a fine-tuning recipe to fix it.

## The tension nobody was measuring

Role-playing agents (RPAs) are LLMs steered via system prompts or fine-tuning to behave like a specific character, persona, or professional archetype. They power everything from interactive fiction and game NPCs to customer-service bots and AI companions. The problem is that alignment training, the RLHF and safety fine-tuning baked into every major model, teaches the model to be helpful, harmless, and honest as a baseline. Those values do not always match the character being played.

[Prior surveys on role-playing LLMs](https://arxiv.org/html/2601.10122v1) have flagged value alignment as an open challenge, noting that current research reveals challenges in character hallucination, point-in-time role-playing, and value alignment. But existing benchmarks mostly test surface-level fidelity: does the model sound like the character? Does it know the character's backstory? What they skip is the harder question: when the character's values and the model's alignment training point in opposite directions, which one wins?

![Comparison of traditional LLM evaluation vs. RoleCDE's scenario-based value conflict evaluation](https://arxiv.org/html/2606.01552v1/Figures/motivation.png) 

The answer, consistently, is alignment wins. The paper calls this **Role Value Decoupling**: the agent systematically defaults to morality-consistent decisions rather than role-specific values when the two conflict, even when the role is explicitly defined in the prompt. This is not a bug in one model. It shows up across every mainstream LLM tested.

## What RoleCDE actually tests

The benchmark is built around **cognitive dilemma scenarios**, structured situations where a character must make a decision that pits their role-specific values against alignment-oriented constraints. Think of it as a moral trolley problem, but the stakes are whether the model stays in character rather than whether it picks the right answer.

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [RoleCDE: Benchmarking and Mitigating Role-Alignment Trade-offs in Role-Playing Agents](https://arxiv.org/abs/2606.01552)
- [Too Good to be Bad: On the Failure of LLMs to Role-Play Villains](https://arxiv.org/html/2511.04962v1)
- [Role-Playing Agents Driven by Large Language Models: Current Status, Challenges, and Future Trends](https://arxiv.org/html/2601.10122v1)
- [RVBench: Role Values Benchmark for Role-Playing LLMs](https://www.sciencedirect.com/science/article/pii/S2949882125000684)
- [Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents](https://arxiv.org/html/2602.13234)