arXiv AI September 18, 2026 By Frank E. Bobe III, Gregory D. Vetaw, Darshan W. Bryner, Matthew G. Cook, Jose L. Salas-Vernis

Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models

Read the original on arXiv AI →

Deep Noir is a framework that autonomously discovers optimal activation‑steering parameters in transformer models by leveraging Logit Lens convergence and causal head‑level attribution. It demonstrates significant performance gains across models ranging from 1B to 9B parameters, improving spam detection by up to 42 percentage points and SST‑2 sentiment classification by 13.1 percentage points without code changes. The approach also reveals that increased steering magnitude expands a predictable prompt‑injection attack surface, highlighting security implications for agent systems using steered classifiers.

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

Jul 14

DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs

arXiv:2607. 11228v1 Announce Type: cross Abstract: While Large Vision-Language Models (LVLMs) demonstrate remarkable capabilities, they remain highly susceptible to embedded social biases.

By Anqi Li, Jie Zhang, Zhongqi Wang, Songkai Xue, Jiahao Wang, Shiguang Shan, Xilin Chen

llms agents reinforcement-learning multimodal benchmarks safety

Hugging Face Trending Papers

Jul 13

DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs

While Large Vision-Language Models (LVLMs) demonstrate remarkable capabilities, they remain highly susceptible to embedded social biases. Existing bias evaluation protocols predominantly rely on static datasets, which provide only a superficial assessment, as their fixed test cases cannot adaptively evolve to measure the true depth and limits of model vulnerabilities.

llms agents reinforcement-learning multimodal benchmarks safety

arXiv AI

Jun 2

Constitutional Black-Box Monitoring for Scheming in LLM Agents

arXiv:2603. 00829v2 Announce Type: replace-cross Abstract: Safe deployment of Large Language Model (LLM) agents in autonomous settings requires reliable oversight mechanisms.

By Simon Storf, Rich Barton-Cooper, James Peters-Gill, Marius Hobbhahn

llms agents

arXiv AI

Jun 19

The Autonomy Tax: Defense Training Breaks LLM Agents

arXiv:2603. 19423v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents increasingly rely on external tools (file operations, API calls, database transactions) to autonomously complete complex multi-step tasks.

By Shawn Li, Yue Zhao

llms agents benchmarks safety

arXiv AI

Jun 16

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw

arXiv:2605. 11047v2 Announce Type: replace-cross Abstract: Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts.

By Hongwei Yao, Yiming Liu, Yiling He, Bingrun Yang

agents robotics benchmarks safety

arXiv Machine Learning

Jun 11

When is Your LLM Steerable?

arXiv:2606. 11599v1 Announce Type: cross Abstract: Activation steering offers a lightweight approach to control language models' behavior at inference time, but whether it succeeds or fails heavily depends on the prompt, concept, model, and steering configuration.

By Chenrui Fan, Yize Cheng, Ming Li, Soheil Feizi, Tianyi Zhou

llms

One email a morning, machine-written

Related stories

DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs

DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs

Constitutional Black-Box Monitoring for Scheming in LLM Agents

The Autonomy Tax: Defense Training Breaks LLM Agents

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw

When is Your LLM Steerable?