arXiv Machine Learning July 13, 2026 By Lucas Pinto

Present but Rescaled: Chat-to-Agent Transfer of Additive Activation Steering

Read the original on arXiv Machine Learning →

arXiv:2607. 09156v1 Announce Type: new Abstract: Additive activation steering (injecting a scaled residual-stream direction during generation) is calibrated almost entirely in single-turn chat, yet the models it targets are increasingly deployed as tool-using ReAct agents.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv AI

Jul 7

Transplanting, inverting, and preventing a misalignment persona: method-conditional emergent misalignment in Qwen2.5

arXiv:2607. 04510v1 Announce Type: cross Abstract: Emergent misalignment (EM) -- the broad misbehaviour a language model acquires after fine-tuning on narrow harmful data -- is mediated in Qwen2.

By Lyndon Drake (University of Oxford), Zandi Eberstadt (University of Oxford)

llms fine-tuning

arXiv Machine Learning

Jul 31

Safety-Gated Agentic Supervisory Control on a Coupled Distillation Benchmark: Regime Map, Auditable Gate, and Co-Design Findings

arXiv:2607. 27849v1 Announce Type: cross Abstract: An open-weight LLM can write composition setpoints every five minutes.

By Christian Rosenthal

llms agents efficiency benchmarks safety

arXiv AI

6d ago

Detecting a Route Flip Is Easier Than Knowing Whether to Fix It: Causal Route-Mediated Damage in Quantized Mixture-of-Experts

arXiv:2608. 11212v1 Announce Type: new Abstract: Top-k Mixture-of-Experts (MoE) routing is discontinuous, so a deployment-motivated numerical disturbance -- simulated 4-bit KV-cache quantization read by a protected BF16 gate -- pushes tokens across decision boundaries and flips which experts fire.

By Parvel Gu

efficiency

arXiv AI

Jul 28

Gubernaut: A Deterministic Homeostatic Controller for Affect-Regulated LLM Agents, Validated Across Independent Model Families

arXiv:2607. 24339v1 Announce Type: new Abstract: Large language model (LLM) agents inherit reactive failure modes: escalation under provocation, sycophantic drift under flattery, perseveration when stuck.

By Dushyant Sharma

llms agents safety

arXiv AI

Jun 15

Order Is Not Control: Driven-Dissipative Response Laws Across Artificial and Biological Systems

arXiv:2606. 12923v2 Announce Type: replace-cross Abstract: AI alignment, interpretability, steering, and neural perturbation studies identify order-inducing objects.

By Gareth Seneque, Lap-Hang Ho, Nafise Erfanian Saeedi, Jeffrey Molendijk, Tim Elson

llms safety

arXiv AI

Jun 12

Order Is Not Control

arXiv:2606. 12923v1 Announce Type: cross Abstract: AI alignment, interpretability, steering, and neural perturbation studies identify order-inducing objects.

By Gareth Seneque, Lap-Hang Ho, Nafise Erfanian Saeedi, Jeffrey Molendijk, Tim Elson

llms safety

Related stories

Transplanting, inverting, and preventing a misalignment persona: method-conditional emergent misalignment in Qwen2.5

Safety-Gated Agentic Supervisory Control on a Coupled Distillation Benchmark: Regime Map, Auditable Gate, and Co-Design Findings

Detecting a Route Flip Is Easier Than Knowing Whether to Fix It: Causal Route-Mediated Damage in Quantized Mixture-of-Experts

Gubernaut: A Deterministic Homeostatic Controller for Affect-Regulated LLM Agents, Validated Across Independent Model Families

Order Is Not Control: Driven-Dissipative Response Laws Across Artificial and Biological Systems

Order Is Not Control