arXiv Computation and Language By Yao Fu, Lijia Huang, Xiaomin Li, Runchao Li, Yu Yin, Kenneth A. Loparo

When Personality Meets Quantization: A Layer-wise MBTI Analysis of Quantized LLMs

Read the original on arXiv Computation and Language →

The paper presents a systematic MBTI analysis of open‑source large language models (LLMs) across various quantization levels, including mainstream 4‑bit and extreme 2‑bit settings. It examines how personality traits emerge layer‑by‑layer through entropy and confidence‑gap dynamics, and introduces Uncertainty‑Amplified Layer Decoding (UALD) to study decoding‑induced personality drift. Findings show that personality is not static but depends on layer, quantization, prompting, and decoding, with ENFJ traits dominating, 4‑bit quantization preserving coarse structure, and 2‑bit quantization disrupting fine‑grained consistency.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI
Aug 19

Beyond BFI: The CSI for Enhanced Reliability and Validity in Evaluating LLM Personality Traits

The paper introduces the Core Sentiment Inventory (CSI), a new personality trait evaluation tool for large language models (LLMs) that addresses reliability and validity issues found in existing methods like the Big Five Inventory (BFI). CSI is designed specifically for LLMs, supports both English and Chinese, and provides detailed psychological portraits of model behavior. Experiments show that CSI captures nuanced behavioral patterns, improves reliability, and correlates strongly (above 0.85) with real-world LLM outputs.

By Huanhuan Ma, Haisong Gong, Xiaoyuan Yi, Xing Xie, Philip S. Yu, Dongkuan Xu
arXiv AI
Aug 26

AgentWorld: Personality-Aware Reliability Evaluation for Agentic Information Retrieval

AgentWorld is a simulation framework that evaluates agentic information retrieval by incorporating diverse user personalities based on the Big Five (OCEAN) traits, stateful tool-use environments, and a pass$^k$ consistency metric with structured fault classification and partial-credit scoring. It includes a risk analyzer that uses Monte‑Carlo rollouts and advanced scoring methods to quantify trajectory brittleness and attack attribution. Experiments with conversational analytics, customer‑support agents, and adversarial stress‑testing demonstrate that personality variation reveals failure modes hidden by uniform testing, such as cross‑domain leakage, contextual drift, and significant quality gaps across personas.

By Gunja Agarwal, Arup Kumar Das, Arun Menon, Jitesh Chandra Mishra, Vignesh Divakaran
arXiv AI
Jul 10

Persona Cartography: Charting Language Model Personality Traits in Weight Space

arXiv:2607. 07916v1 Announce Type: new Abstract: Large language models exhibit recurring behavioural patterns -- personas -- that shape generalisation and safety, but we lack reliable tools for decomposing, measuring, and controlling them.

By Luke Baines, Anton Gonzalvez Hawthorne, Mariia Koroliuk, Irakli Shalibashvili, Cl\'ement Dumas, Konstantinos Voudouris, David Demitri Africa