Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle
원문 초록
Recent innovations in language model training have demonstrated that it is possible to create highly performant models that are small enough to run on a smartphone. As these models are deployed in an increasing number of domains, it is critical to ensure that they are aligned with human preferences and safety considerations. In this report, we present our methodology for safety aligning the Phi-3 series of language models. We utilized a "break-fix" cycle, performing multiple rounds of dataset curation, safety post-training, benchmarking, red teaming, and vulnerability identification to cover a variety of harm areas in both single and multi-turn scenarios. Our results indicate that this approach iteratively improved the performance of the Phi-3 models across a wide range of responsible AI benchmarks. Finally, we include additional red teaming strategies and evaluations that were used to test the safety behavior of Phi-3.5-mini and Phi-3.5-MoE, which were optimized for multilingual capabilities.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
mTSBench: Benchmarking Multivariate Time Series Anomaly Detection and Model Selection at Scale
BMFM-RNA: An Open Framework for Building and Evaluating Transcriptomic Foundation Models
MLAR: Multi-layer Large Language Model-based Robotic Process Automation Applicant Tracking
A Systematic Analysis of Hybrid Linear Attention
Inaugural MOASEI Competition at AAMAS'2025: A Technical Report
Finance Language Model Evaluation (FLaME)
토론
이 논문과 연결된 포럼 글이 아직 없습니다.