Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle
Abstract
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.
Community
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- DataComp-LM: In search of the next generation of training sets for language models (2024)
- Safety Arithmetic: A Framework for Test-time Safety Alignment of Language Models by Steering Parameters and Activations (2024)
- ASTPrompter: Weakly Supervised Automated Language Model Red-Teaming to Identify Likely Toxic Prompts (2024)
- SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Models (2024)
- DistillSeq: A Framework for Safety Alignment Testing in Large Language Models using Knowledge Distillation (2024)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 31
Browse 31 models citing this paperDatasets citing this paper 0
No dataset linking this paper