AI ethics evidence index Search
Source-linked daily record
Evidence record 448 · automatically gathered

EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures

LLM evaluation and AI safety face a shared measurement problem: benchmark scores, reward-model signals, and reported safety metrics can improve while the latent properties they are meant to represent remain difficult to verify. This paper combines a hybrid survey - a systematic search paired with narrative synthesis and separately tracked grey evidence - with a conceptual framework and a structured ten-model audit. The synthesis spans eight evidence streams: benchmark validity, dynamic evaluatio

Record details

Published: 29 June 2026
Source: arXiv
Category: Research
Topics: Safety & alignment · Transparency
Retrieved: 14 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: Safety & alignment · Shared topic: Transparency

FLARE-AI: Flaw Reporting for AI

arXiv · 30 June 2026

Shared topic: Safety & alignment · Shared topic: Transparency

FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning

arXiv · 30 June 2026

Shared topic: Safety & alignment · Shared topic: Transparency

Bridging Vision and Language Concepts through Optimal Transport Semantic Flow

arXiv · 25 June 2026

Shared topic: Safety & alignment · Shared topic: Transparency

NebulaExp-8B: An Empirical Post-Training Pipeline via Full-Scale Ablation Research

arXiv · 25 June 2026

Shared topic: Safety & alignment · Shared topic: Transparency

Explainable Reinforcement Learning for Adaptive Traffic Signal Control

arXiv · 4 July 2026

Shared topic: Safety & alignment · Shared topic: Transparency

Phoneme-Level Mispronunciation Screening in Polish-Speaking Children with an Explainable Assistant

arXiv · 23 June 2026

How to cite this record

ethics.ai (29 June 2026), “EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures,” evidence record 448, https://ethics.ai/record/448 (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.