Papers
arxiv:2505.19766

SGM: A Framework for Building Specification-Guided Moderation Filters

Published on May 26
Authors:
,
,

Abstract

SGM is a flexible framework for training moderation filters based on user-defined specifications, automating data generation and supporting diverse alignment goals.

AI-generated summary

Aligning large language models (LLMs) with deployment-specific requirements is critical but inherently imperfect. Despite extensive training, models remain susceptible to misalignment and adversarial inputs such as jailbreaks. Content moderation filters are commonly used as external safeguards, though they typically focus narrowly on safety. We introduce SGM (Specification-Guided Moderation), a flexible framework for training moderation filters grounded in user-defined specifications that go beyond standard safety concerns. SGM automates training data generation without relying on human-written examples, enabling scalable support for diverse, application-specific alignment goals. SGM-trained filters perform on par with state-of-the-art safety filters built on curated datasets, while supporting fine-grained and user-defined alignment control.

Community

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2505.19766 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2505.19766 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2505.19766 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.