# AEGIS: Awareness-Enhanced Guidance for Iterative Safeguard

arXiv: https://arxiv.org/abs/2607.17713 · HTML: https://mlsanity.com/paper/2607.17713/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 18 references that are in the map.
- First posted on arXiv 20 Jul 2026.

## Builds on (most cited first, 10 of 10)

- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [Deep Learning for Text Style Transfer: A Survey](https://mlsanity.com/paper/2011.00416/) (2020)
- [BEEP! Korean Corpus of Online News Comments for Toxic Speech Detection](https://mlsanity.com/paper/2005.12503/) (2020)
- [A Review of Text Style Transfer using Deep Learning](https://mlsanity.com/paper/2109.15144/) (2021)
- [Proactively Reducing the Hate Intensity of Online Posts via Hate Speech Normalization](https://mlsanity.com/paper/2206.04007/) (2022)
- [K-MHaS: A Multi-label Hate Speech Detection Dataset in Korean Online News Comment](https://mlsanity.com/paper/2208.10684/) (2022)
- [DetoxLLM: A Framework for Detoxification with Explanations](https://mlsanity.com/paper/2402.15951/) (2024)
- [STATE ToxiCN: A Benchmark for Span-level Target-Aware Toxicity Extraction in Chinese Hate Speech Detection](https://mlsanity.com/paper/2501.15451/) (2025)
- [Target Span Detection for Implicit Harmful Content](https://mlsanity.com/paper/2403.19836/) (2024)
- [K-HATERS: A Hate Speech Detection Corpus in Korean with Target-Specific Ratings](https://mlsanity.com/paper/2310.15439/) (2023)

## Abstract (arXiv)

Span-level rationales are often assumed to improve controllability in text detoxification, but it remains unclear when such guidance helps and when it introduces trade-offs. We present Awareness-Enhanced Guidance for Iterative Safeguard (AEGIS) as an exploratory framework for studying span-guided multilingual detoxification across English, Mandarin Chinese, and Korean. AEGIS combines span-level detector outputs with frozen generator backbones, allowing harmful spans, intensity labels, and target attributes to be provided as structured guidance during rewriting. Rather than claiming state-of-the-art detoxification performance, we analyze how span guidance affects the balance between toxicity reduction and meaning preservation across generator families, model scales, and languages. Our results suggest that span-guided detoxification is conditionally useful: explicit rationales change the trade-off between toxicity reduction and meaning preservation, but their effects depend strongly on the generator backbone and the linguistic context. These findings highlight both the promise and the limitations of span-level control signals for multilingual detoxification.