AI ethics evidence index Search
Source-linked daily record
Evidence record 11968 · automatically gathered

How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?

Modern LLMs are alarmingly susceptible to surprisingly simple immaterial changes of input prompts: a casual hint, an incorrectly labeled few-shot example, or a fake prior assistant turn often flips an originally correct answer. We study where this susceptibility, spanning sycophancy and related cue-induced biases, lives inside the model. Across five model families and seven BCT bias types, we extract a per-bias direction from hidden states and triangulate it through three measures: probing, leav

Record details

Published: 20 July 2026
Source: arXiv
Category: Research
Topics: Bias & fairness · Safety & alignment
Retrieved: 21 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Sympathetic Framing: Evaluating AI Alignment across Sociodemographic Groups

arXiv cs.CL (ethics-relevant NLP) · 22 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

PG-KINN: A Physics-Informed Petrov-Galerkin Kolmogorov-Arnold Network for Solving Forward and Inverse PDEs

arXiv cs.LG · 22 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Language Models Embody and Amplify Human Cognitive Distortions: What Is to Be Done?

arXiv · 22 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Spectral Prior for Reducing Exposure Bias in Diffusion Models

HuggingFace Daily Papers · 23 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling

arXiv · 17 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

AI Alignment Amplifies the Role of Race, Gender, and Disability in Hiring Decisions

arXiv cs.CY · 16 July 2026

How to cite this record

ethics.ai (20 July 2026), “How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?,” evidence record 11968, https://ethics.ai/record/11968 (originally published by arXiv).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.