LLM Digest

Story

arxiv_llm_reliability · Nov 9, 2026 · paper

Source brief

Vendi Anomaly Scores for Efficient and Accurate Anomaly Detection

arxiv.orgNov 9, 2026
original source linked

In brief

Anomaly detection (AD) is a critical for trustworthy machine learning and scientific discovery. Existing methods typically define anomalies through local density, isolation heuristics, or distance from a fitted distri...

Continues in

Anomaly Detection — open the evidence trace →

Read the original at arxiv.org →Open in live feed Read that day’s brief

Vendi Anomaly Scores for Efficient and Accurate Anomaly Detection

Earlier in this thread 3 items

MoG-VLM: Efficient Video Anomaly Detection through Motion-Guided Vision-Language Models

LogCDM: Interpretable Log Anomaly Detection via Behavioral Graph Drift and Adaptive Signal Fusion

Flownomaly: Single-Step Conditional-Mean Recovery for Few-Shot Anomaly Detection