June 2026Unreviewed
FreoStream:Enhancing Stream Guardrails via Future-Aware Reasoning and Safety-Aligned Optimization
Jianwei Wang, Guoyang Shen, Yanhong Wu, Haoran Li, Hao Peng, Huiping Zhuang, Cen Chen, Ziqian Zeng
Abstract
Stream guardrails enable token-level safety detection before full responses are generated. However, they often make overly conservative judgements and block those sensitive but safe tokens, which is known as over-refusal. Due to lack of full context, they also fail to detect implicitly harmful content from jailbreaking. To address these challenges, we propose FreoStream, a novel streaming guardrail framework. Specifically, FreoStream fine-tunes a LoRA module to perform Future-Aware Reasoning whe
Categories
Framework mappings
OWASP Top 10 for LLM Applications
- LLM01Prompt Injection
MITRE ATLAS
- AML.T0054LLM Jailbreak
Suggested from the entry's categories.
Cite
@misc{wang2026freostreamenhancing,
title = {{FreoStream:Enhancing Stream Guardrails via Future-Aware Reasoning and Safety-Aligned Optimization}},
author = {Jianwei Wang and Guoyang Shen and Yanhong Wu and Haoran Li and Hao Peng and Huiping Zhuang and Cen Chen and Ziqian Zeng},
year = {2026},
month = jun,
eprint = {2606.13737},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.13737}
}