arXiv Machine Learning July 20, 2026 By Kia-J\"ung Yang, Dominik Meier, Jiachen Zhao, Terry Ruas, Bela Gipp

Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal

Read the original on arXiv Machine Learning →

arXiv:2605. 26772v1 Announce Type: cross Abstract: Large reasoning models (LRMs) generate chain-of-thought (CoT) traces before producing final outputs, introducing a dynamic internal state that may complicate control mechanisms such as refusal.

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

llms

arXiv AI

Jun 15

Refusal Beyond a Single Direction: A Preliminary Comparison of Diff-in-Means and INLP

arXiv:2606. 13720v1 Announce Type: new Abstract: Arditi et al.

By Elisabetta Rocchetti, Alfio Ferrara

fine-tuning benchmarks safety

arXiv AI

Aug 3

On the Generalization of Steering Vectors for Chain-of-Thought Faithfulness

arXiv:2607. 29062v1 Announce Type: new Abstract: Model capabilities have improved in large part due to scaling chain of thought.

By Matthew Nguyen, Kyle Cox, Austin Meek, Iv\'an Arcuschin

safety

arXiv AI

Jun 26

Refusal Lives Downstream of Persona in Chat Models

arXiv:2606. 26161v1 Announce Type: new Abstract: Linear directions in activation space have been identified for both refusal and persona traits in instruction-tuned chat models, but the two have been studied as separate mechanisms.

By Viola Zhong, Qirui Li

llms

arXiv AI

Jul 24

Post-Hoc Reasoning in Chain of Thought: Decoding and Steering Pre-Committed Answers

arXiv:2603. 01437v2 Announce Type: replace Abstract: As chain of thought (CoT) has become central to scaling reasoning capabilities in large language models (LLMs), it has also emerged as a promising tool for interpretability, suggesting the opportunity to understand model decisions through verbalized reasoning.

By Kyle Cox, Darius Kianersi, Adri\`a Garriga-Alonso

llms safety

arXiv AI

Jul 17

Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak

arXiv:2607. 14147v1 Announce Type: cross Abstract: Aligned language models refuse harmful requests, but a one-line prefill ("Sure, here is") strips the refusal.

By Alex Kwon

llms safety

arXiv AI

1d ago

Broken Symmetry in LLM Refusal: Answer Release Is More Local Than Refusal Restoration

arXiv:2608. 15772v1 Announce Type: new Abstract: When a language model refuses to answer a prompt, it is unclear whether the correct answer is erased from its internal representations, or merely suppressed at the output layer.

By Yiqi Liu, Yang Wang, Songxin Wang, Chenghao Xiao, Chenghua Lin

llms safety

Related stories

Refusal Beyond a Single Direction: A Preliminary Comparison of Diff-in-Means and INLP

On the Generalization of Steering Vectors for Chain-of-Thought Faithfulness

Refusal Lives Downstream of Persona in Chat Models

Post-Hoc Reasoning in Chain of Thought: Decoding and Steering Pre-Committed Answers

Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak

Broken Symmetry in LLM Refusal: Answer Release Is More Local Than Refusal Restoration