arXiv AI October 28, 2026 By Chuming Wei, Churui Yang, Zihan Zhang, Ruien Li, Wusheng Zhang, Yan Zhang, Guangwen Yang

ADTGuard: Adaptive Depth-Time Safeguard for Streaming Inference in Large Language Models

ADTGuard is a new safeguard for large language models that monitors each token before it is streamed to users. It uses Depth-Time Safety Dynamics to combine safety signals from multiple layers and a Causal Depth Router to select the most relevant layers for each prefix. With only response‑level labels, it refines token‑level training targets and achieves strong streaming detection performance with minimal parameters and latency.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jun 10

Stop Early, Spend Less: Hidden-State Probes as a Practical Recipe for Streaming Moderation of LLM Outputs

arXiv:2606. 10487v1 Announce Type: cross Abstract: Deploying large language models in user-facing systems requires efficient output safety filtering.

By Huizhen Shu, Xuying Li, Piao Xue

llms safety

arXiv AI

Sep 1

SingProbe Technical Report

arXiv:2608.30703v1 Announce Type: cross Abstract: Runtime guardrails are essential for reliable large language model (LLM) deployment, yet existing approaches typically rely on independent, external...

By Sing Team

llms benchmarks safety

Hugging Face Trending Papers

Jun 1

SentGuard: Sentence-Level Streaming Guardrails for Large Language Models

Large language models increasingly stream long, reasoning-intensive responses in real time, making when to moderate as critical as whether to moderate. Existing guardrails fall into two unsatisfactory extremes: response-level methods delay intervention until the full output is generated, whereas token-level methods act on incomplete semantics, often producing unstable decisions and excessive guard invocations.

llms benchmarks safety

arXiv Machine Learning

Jun 4

Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning

arXiv:2603. 07445v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often require fine-tuning (FT) to perform well on downstream tasks, but FT can induce safety-alignment drift even when the training dataset contains only benign data.

By Guoli Wang, Haonan Shi, Tu Ouyang, An Wang

llms fine-tuning safety

arXiv AI

Jun 19

SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling

arXiv:2606. 19755v1 Announce Type: cross Abstract: Speculative inference accelerates large language model (LLM) decoding but provides no inherent safety guarantees.

By Haotian Xu, Zeyang Zhang, Linbao Li, Huadi Zheng, Yu Li, Cheng Zhuo

llms benchmarks safety

arXiv AI

1d ago

UpSafe$^\circ$C: Upcycling for Controllable Safety in Large Language Models

UpSafe°C is a unified framework that enhances large language model safety by upcycling safety‑critical layers into a sparse Mixture‑of‑Experts structure, with a router acting as a soft guardrail. It employs a two‑stage supervised fine‑tuning strategy to improve safety discrimination while preserving general capabilities, and introduces a safety temperature mechanism for dynamic inference‑time control of the safety‑utility trade‑off. Experiments across multiple benchmarks and model scales show robust safety improvements against harmful and jailbreak inputs while maintaining competitive general performance.

By Yuhao Sun, Zhuoer Xu, Shiwen Cui, Kun Yang, Lingyun Yu, Yongdong Zhang, Hongtao Xie

llms fine-tuning benchmarks safety

One email a morning, machine-written

Related stories

Stop Early, Spend Less: Hidden-State Probes as a Practical Recipe for Streaming Moderation of LLM Outputs

SingProbe Technical Report

SentGuard: Sentence-Level Streaming Guardrails for Large Language Models

Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning

SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling

UpSafe$^\circ$C: Upcycling for Controllable Safety in Large Language Models