5 papers
Architectural Wisdom: A Framework for Governing Optimization in AI Systems
Edward Y. Chang
Modern AI systems exhibit structural failures that capability scaling alone does not reliably fix: they optimize under-specified objectives with no architectural mechanism to quest…
Diagnosing and Mitigating Sycophancy and Skepticism in LLM Causal Judgment
Edward Y. Chang
Large language models increasingly fail in a way that scalar accuracy cannot diagnose: they produce a sound reasoning trace and then abandon it under social pressure or an authorit…
Internal Reasoning vs. External Control: A Thermodynamic Analysis of Sycophancy in Large Language Models
Edward Y. Chang
Large Language Models exhibit sycophancy: prioritizing agreeableness over correctness. Current remedies evaluate reasoning outcomes: RLHF rewards correct answers, self-correction c…
Modeling Emotions and Ethics with Large Language Models
Edward Y. Chang
This paper explores the integration of human-like emotions and ethical considerations into Large Language Models (LLMs). We first model eight fundamental human emotions, presented…
Integrating Emotional and Linguistic Models for Ethical Compliance in Large Language Models
Edward Y. Chang
This research develops advanced methodologies for Large Language Models (LLMs) to better manage linguistic behaviors related to emotions and ethics. We introduce DIKE, an adversari…