paper-with-me

Papers

Policy-Grounded Safety Evaluation of 20 Large Language Models

2025-07-19 · Juan Manuel Contreras arxiv

As large language models (LLMs) become increasingly integrated into real-world applications, scalable and rigorous safety evaluation is essential. This paper introduces Aymara AI, a programmatic platform for generating and administering customized, policy-grounded safety evaluations. Aymara AI transforms natural-language safety policies into adversarial prompts and scores model responses using an AI-based rater validated against human judgments. We demonstrate its capabilities through the Aymara LLM Risk and Responsibility Matrix, which evaluates 20 commercially available LLMs across 10 real-world safety domains. Results reveal wide performance disparities, with mean safety scores ranging from 86.2% to 52.4%. While models performed well in well-established safety domains such as Misinformation (mean = 95.7%), they consistently failed in more complex or underspecified domains, notably Privacy & Impersonation (mean = 24.3%). Analyses of Variance confirmed that safety scores differed significantly across both models and domains (p < .05). These findings underscore the inconsistent and context-dependent nature of LLM safety and highlight the need for scalable, customizable tools like Aymara AI to support responsible AI development and oversight.

📄 PDF Abstract BibTeX arXiv:2507.14719

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models

2026-05-01 · Yunhan Zhao, Zhaorun Chen, Xingjun Ma, Yu-Gang Jiang 외 arxiv

As Large Language Models (LLMs) are increasingly deployed in cross-linguistic contexts, ensuring safety in diverse regulatory and cultural environments has become a critical challenge. However, existing multilingual benc…

Machine Translation

UbuntuGuard: A Culturally-Grounded Policy Benchmark for Equitable AI Safety in African Languages

2026-01-19 · Tassallah Abdullahi, Macton Mgonzo, Mardiyyah Oduwole, Paul Okewunmi 외 arxiv

Current guardian models are predominantly Western-centric and optimized for high-resource languages, leaving low-resource African languages vulnerable to evolving harms, cross-lingual failures, and cultural misalignment.…

Cross-Lingual Transfer

Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety

2026-06-26 · Ting Ma, Xiufeng Huang, Benlei Cui, Xiaowen Xu 외 arxiv

As large language models are increasingly deployed in real-world systems, safety failures can still lead to harmful outputs and dangerous misuse. We argue that the essence of safety is adversarial: many failures arise no…

Adversarial RobustnessReinforcement Learning

CourtGuard: A Model-Agnostic Framework for Zero-Shot Policy Adaptation in LLM Safety

2026-02-26 · Umid Suleymanov, Rufiz Bayramov, Suad Gafarli, Seljan Musayeva 외 arxiv

Current safety mechanisms for Large Language Models (LLMs) rely heavily on static, fine-tuned classifiers that suffer from adaptation rigidity, the inability to enforce new governance rules without expensive retraining. …

A Self-Improving Architecture for Dynamic Safety in Large Language Models

2025-11-10 · Tyler Slater arxiv

Context: Large Language Models (LLMs) rely on static, pre-deployment safety mechanisms that cannot adapt to adversarial threats discovered after release. Objective: To design a software architecture enabling LLM-based sy…