mlsanityData to 14 Oct 2026

When Roles Fail: Epistemic Constraints on Advocate Role Fidelity in LLM-Based Political Statement Analysis

arXiv 2604.27228 ↗First posted 29 Apr 2026cs.AI cs.CL cs.CYcited by 2

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Democratic discourse analysis systems increasingly rely on multi-agent LLM pipelines in which distinct evaluator models are assigned adversarial roles to generate structured, multi-perspective assessments of political statements. A core assumption is that models will reliably maintain their assigned roles. This paper provides the first systematic empirical test of that assumption using the TRUST pipeline. We develop an epistemic stance classifier that identifies advocate roles from reasoning text without relying on surface vocabulary, and measure role fidelity across 60 political statements (30 English, 30 German) using four metrics: Role Drift Index (RDI), Expected Drift Distance (EDD), Directional Drift Index (DDI), and Entropy-based Role Stability (ERS). We identify two failure modes - the Epistemic Floor Effect (fact-check results create an absolute lower bound below which the legitimizing role cannot be maintained) and Role-Prior Conflict (training-time knowledge overrides role instructions for factually unambiguous statements) - as manifestations of a single mechanism: Epistemic Role Override (ERO). Model choice significantly affects role fidelity: Mistral Large outperforms Claude Sonnet by 28pp (67% vs. 39%) and exhibits a qualitatively different failure mode - role abandonment without polarity reversal - compared to Claude's active switch to the opposing stance. Role fidelity is language-robust. Fact-check provider choice is not universally neutral: Perplexity significantly reduces Claude's role fidelity on German statements (Delta = -15pp, p = 0.007) while leaving Mistral unaffected. These findings have direct implications for multi-agent LLM validation: a system validated without role fidelity measurement may systematically misrepresent the epistemic diversity it was designed to provide.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 7 of its 60 references that are in the map. First posted on arXiv 29 Apr 2026.

Builds on

7 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2026Peer Identity Bias in Multi-Agent LLM Evaluation: An Empirical Study Using the TRUST Democratic Discourse Analysis Pipeline24 Apr
  2. 2026From Safety Risk to Design Principle: Peer-Preservation in Multi-Agent LLM Systems and Its Implications for Orchestrated Democratic Discourse Analysis9 Apr
  3. 2025Revealing Political Bias in LLMs through Structured Multi-Agent Debate13 Jun

Other references · 4

  1. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  2. 2025Consistently Simulating Human Personas with Multi-Turn Reinforcement Learningcited by 39
  3. 2025Enhancing Persona Consistency for LLMs' Role-Playing using Persona-Aware Contrastive Learningcited by 27
  4. 2024Generative Adversarial Reviews: When LLMs Become the Criticcited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Can Multi-Agent LLMs Identify Their Peers? Stylometric Fingerprinting in Role-Constrained Political Analysiscited by 1
  2. 2026Beyond Acoustic Emotion Recognition: Multimodal Pathos Analysis in Political Speech Using LLM-Based and Acoustic Emotion Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024GermanPartiesQA: Benchmarking Commercial Large Language Models and AI Companions for Political Alignment and Sycophancy
  2. 2026Validating Political Position Predictions of Arguments
  3. 2025Examining Alignment of Large Language Models through Representative Heuristics: The Case of Political Stereotypes
  4. 2025PReSS: An Automated Black-Box Framework for Evaluating Political Stance Stability in LLMs
  5. 2026Belief Engine: Configurable and Inspectable Stance Dynamics in Multi-Agent LLM Deliberation
  6. 2026Auditing Political Alignment in LLM Assistants: Engagement, Stance, and User Identity
  7. 2025From Delegates to Trustees: How Optimizing for Long-Term Interests Shapes Bias and Alignment in LLM
  8. 2025Community-Aligned Behavior Under Uncertainty: Evidence of Epistemic Stance Transfer in LLMs