Signals 4 · free daily AI digest

SpecGuard: Inference-Time Backdoor Detection For Free

Paper recorded by Signals 4 on 2026-09-10 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-10 on arXiv · recorded by Signals 4 on 2026-09-11

Category: cs.CL · 自然语言处理 · first seen 2026-09-11

Abstract

Large language models are often fine-tuned, shared, or downloaded from third parties, so a deployed model may carry a hidden backdoor that behaves normally on benign inputs but switches to attacker-controlled behavior when a secret trigger appears. While backdoors can be audited before deployment, runtime monitoring remains important for models that are frequently updated. The challenge is that LL

Read on arXiv →

#72 most recent of 186 cs.CL papers we have recorded · ↑ newer: Target leakage, not model class, explains reported accuracy in survey- · ↓ older: Whisper-Based Speech Transcription from Videos Across Multiple Languag
Cite this page: SpecGuard: Inference-Time Backdoor Detection For Free: the #72 most recent of 186 cs.CL papers we have recorded (as of 2026-09-10). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/specguard-inference-time-backdoor-detection-for-free.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CL papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions