Production noise creates correlated perturbations that act as shared evidence, causing human groups to form and maintain consensus on incorrect estimates more than comprehension noise or no noise.
arXiv preprint arXiv:2505.21588 , year=
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 3
citation-polarity summary
roles
background 3representative citing papers
FlowSteer is a prompt-only attack that biases multi-agent LLM workflow planning to propagate malicious signals, raising success rates by up to 55%, with FlowGuard as an input-side defense reducing it by up to 34%.
Large language models exhibit normative conformity in addition to informational conformity, and subtle social context can direct which group they conform to.
A systematic audit of LLM-based AI societies finds that 89.7% of 39 studies violate at least one of six PIMMUR validity principles, with reproductions showing that many claimed collective behaviors disappear when controls are tightened.
Peer agreement misleads initially correct LLMs more than it corrects initially wrong ones, with authority labels biasing choices independently of accuracy and reasoning prompts failing to mitigate the asymmetry.
The paper introduces a three-source decomposition showing that answer flips in multi-agent LLM debate include 37% spontaneous instability and 29% harmful conformity, with even vacuous reasoning persuading 20-39% of resistant agents and interventions reducing harmful conformity by 13.6 points.
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
Frontier multimodal models judge Chinese short-video misinformation inconsistently, and their veracity ratings shift when videos carry verified or authoritative channel identities.
The authors introduce agentic microphysics and generative safety to link local agent interactions to population-level risks in agentic AI through a causally explicit framework.
citing papers explorer
-
Private Noise and Public Error in Collective Information Acquisition
Production noise creates correlated perturbations that act as shared evidence, causing human groups to form and maintain consensus on incorrect estimates more than comprehension noise or no noise.
-
FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems
FlowSteer is a prompt-only attack that biases multi-agent LLM workflow planning to propagate malicious signals, raising success rates by up to 55%, with FlowGuard as an input-side defense reducing it by up to 34%.
-
Large Language Models Exhibit Normative Conformity
Large language models exhibit normative conformity in addition to informational conformity, and subtle social context can direct which group they conform to.
-
The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies
A systematic audit of LLM-based AI societies finds that 89.7% of 39 studies violate at least one of six PIMMUR validity principles, with reproductions showing that many claimed collective behaviors disappear when controls are tightened.
-
Easier to Mislead Than to Correct: Harmful and Beneficial Revision in LLM Conformity
Peer agreement misleads initially correct LLMs more than it corrects initially wrong ones, with authority labels biasing choices independently of accuracy and reasoning prompts failing to mitigate the asymmetry.
-
Not All Flips Are Conformity: Decomposing Stance Convergence in Multi-Agent LLM Debate
The paper introduces a three-source decomposition showing that answer flips in multi-agent LLM debate include 37% spontaneous instability and 29% harmful conformity, with even vacuous reasoning persuading 20-39% of resistant agents and interventions reducing harmful conformity by 13.6 points.
-
When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
-
Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation
Frontier multimodal models judge Chinese short-video misinformation inconsistently, and their veracity ratings shift when videos carry verified or authoritative channel identities.
-
Agentic Microphysics: A Manifesto for Generative AI Safety
The authors introduce agentic microphysics and generative safety to link local agent interactions to population-level risks in agentic AI through a causally explicit framework.