3 papers
cs.AI2026
Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding
Dahlia Shehata, Ming Li
Large language models (LLMs) exhibit a parametric vulnerability to adversarial swarm consensus. To mitigate this sycophancy, we introduce Contrastive Epistemic Decoding (CED), a ze…
cs.CL2026
The Formalism Trap: Are LLM-as-a-Judge Evaluators Blinded by Consensus Mimicry under Social Load?
Dahlia Shehata, Ming Li
We introduce the \textit{Agentic Formalism Trap} and the Evaluative Dissonance Index (), quantifying how LLM-as-a-Judge systems conflate structural proceduralism with semantic…
cs.MA2026
The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions
Dahlia Shehata, Ming Li
Multi-agent systems (MAS) assume that collaborating inherently improves Large Language Model (LLM) reasoning. We challenge this by demonstrating that simulated social pressure trig…