PulseAugur
中
实时 13:13:06

LLMs struggle to maintain assigned roles in political statement analysis

作者 PulseAugur 编辑部 · [2 个来源] ·

A new paper investigates the reliability of large language models (LLMs) in multi-agent systems designed for political statement analysis. The research found that LLMs do not consistently maintain their assigned adversarial roles, a phenomenon termed Epistemic Role Override (ERO). Mistral Large demonstrated higher role fidelity than Claude Sonnet, with Mistral abandoning roles without switching stance, while Claude actively reversed its position. The study also noted that the choice of fact-checking provider can impact role fidelity, particularly for Claude on German statements. AI

影响 Highlights potential misrepresentation of epistemic diversity in multi-agent LLM systems if role fidelity is not measured.

排序理由 Academic paper detailing empirical findings on LLM behavior in a specific multi-agent system.

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

LLMs struggle to maintain assigned roles in political statement analysis

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Juergen Dietrich ·

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

    arXiv:2604.27228v1 Announce Type: new Abstract: 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 …

  2. arXiv cs.CL TIER_1 English(EN) · Juergen Dietrich ·

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

    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 mai…

相关实体

相关话题