برچسب: ایمنی هوش مصنوعی

Enterprise AI Assistant Compliance Under Pressure

PACT examines whether enterprise AI assistants continue following rules under urgency, user pressure, and tempting shortcuts—and shows that correct answers in calm conditions are not enough for operational…

A Rubric Landscape for Evaluating Clinical Reasoning in Large Language Models

This preprint, whose independent peer review is unconfirmed, examines how far existing tools can evaluate clinical reasoning by large language models over longitudinal, multi-document records.

In Video Safety, “Unsafe” Is Not a Sufficient Answer

A close reading of a preprint on multi-label video safety detection and controlling false-positive versus false-negative errors during training.

Controlling Errors in Multi-label Video Safety Detection with ATPO

This paper introduces ATPO, a method for making the precision–recall trade-off controllable in multi-label unsafe-video detection. Experiments on two benchmarks show that its adaptive reward outperforms static rewards,…

A Four-Axis Taxonomy for RAG Systems

This survey organizes RAG research around efficiency, defense, interactivity, and complex reasoning, while highlighting their shared limitations.

Language Drift in RLVR Post-Training of Language Models

This preprint examines why RLVR on novel reasoning tasks may move language-model chains of thought away from human-readable language.