AI ethics evidence index Search
Source-linked daily record
Evidence record 16650 · automatically gathered

Item Response Theory for AI Safety

Language models differ in how safely they behave and these differences are measured by safety benchmarks. But aggregated benchmark scores are hard to trust and interpret, because benchmarks duplicate one another, correlate heavily, and models may sandbag when they detect evaluation. To address these issues, we draw on Item Response Theory (IRT), a statistical toolkit for measuring these latents from performance on items with inferred psychometric properties. We fit IRT models to eight safety ben

Record details

Published: 5 August 2026
Source: arXiv
Category: Research
Topics: Safety & alignment
Retrieved: 6 August 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: Safety & alignment

Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming

arXiv red teaming query · 5 August 2026

Shared topic: Safety & alignment

Rectifying Geometric Misalignment: Online Source-Free Adaptation for Class-Imbalanced EEG

arXiv cs.LG · 5 August 2026

Shared topic: Safety & alignment

Scaling Inherently Interpretable Language Models

HuggingFace Daily Papers · 5 August 2026

Shared topic: Safety & alignment

Decoupling Perception from Description: Computation-Grounded Representation Alignment between Multivariate Time Series and Language

arXiv cs.LG · 5 August 2026

Shared topic: Safety & alignment

Mood Matters: How Syntactic Sensitivity Undermines Safety Alignment

arXiv red teaming query · 5 August 2026

Shared topic: Safety & alignment

RepoProbe: Benchmarking Architecture-Aware Repository Comprehension with Checklists

arXiv cs.AI · 5 August 2026

How to cite this record

ethics.ai (5 August 2026), “Item Response Theory for AI Safety,” evidence record 16650, https://ethics.ai/record/16650 (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.