Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements

By Jiawen Deng, Hao Sun, Zhexin Zhang, Jiale Cheng, Minlie Huang. 2023

26 citations recorded by the source corpus.

As generative large model capabilities advance, safety concerns become more pronounced in their outputs. To ensure the sustainable growth of the AI ecosystem, it's imperative to undertake a holistic evaluation and refinement of associated safety risks. This survey presents a framework for safety research pertaining to large models, delineating the landscape of safety risks as well as safety evaluation and improvement methods. We begin by introducing safety issues of wide concern, then delve into safety evaluation methods for large models, encompassing preference-based testing, adversarial attack approaches, issues detection, and other advanced evaluation methods. Additionally, we explore the strategies for enhancing large model safety from training to deployment, highlighting cutting-edge safety approaches for each stage in building large models. Finally, we discuss the core challenges in advancing towards more responsible AI, including the interpretability of safety mechanisms, ongoing safety issues, and robustness against malicious attacks. Through this survey, we aim to provide clear technical guidance for safety researchers and encourage further study on the safety of large models.

Subfields: Information Systems, Safety, Risk, Reliability and Quality, Software, Large Language Models, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Safety, Risk, Reliability and Quality

Binh Tang

In Software

In Large Language Models

In Safety, Robustness & Interpretability

Citations in the corpus

Llama 2: Open Foundation and Fine-Tuned Chat Models (2023, background): “Deng et al. (2023) proposes a taxonomic framework to tackle these issues, and Bergman et al. (2022) delves into the balance between potential positive and negative impacts from releasing dialogue models.”
Using In-Context Learning to Improve Dialogue Safety (2023, background): “While these models are capable of engaging in interesting and coherent dialogue, recent work has shown these systems are prone to generating unsafe content (Xu et al. 2021b; Dinan et al. 2022; Deng et al. 2023; inter alia ).”