arXiv Machine Learning July 24, 2026 By Mohammed Suhail B Nadaf

Emergent Misalignment Recruits a Pre-existing Persona Subspace

Read the original on arXiv Machine Learning →

arXiv:2607. 21356v1 Announce Type: new Abstract: Fine-tuning an aligned language model on a narrow stream of bad advice can make it broadly misaligned on questions unrelated to the training data, a phenomenon called emergent misalignment.

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 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 AI

Jul 1

Evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment

arXiv:2606. 31591v1 Announce Type: cross Abstract: Emergent misalignment (EM) is a recently discovered phenomenon in LLMs where fine-tuning on a narrow misaligned task, such as writing insecure code, leads to broadly misaligned behaviour on unrelated prompts.

By Jason R. Brown, Patrick Leask, Lev McKinney

llms fine-tuning safety

arXiv Machine Learning

Jul 30

Constitutional Midtraining: Content Presence Drives Alignment Gains

arXiv:2607. 26654v1 Announce Type: cross Abstract: Post-training alignment is often shallow, eroding under fine-tuning.

By Desiree Cho, Cameron Tice, Bernie Hogan, Hunar Batra, Puria Radmard, Jun Zhao, Nigel Shadbolt

fine-tuning benchmarks safety

arXiv AI

Jul 29

Similar Models Learn Differently: Final-Window Pretraining Shapes Post-Training Beyond SFT

arXiv:2607. 25063v1 Announce Type: new Abstract: Developers judge a model checkpoint by how it behaves.

By Cen Lu, Yung-Chen Tang, Andrea Cavallaro

reinforcement-learning fine-tuning benchmarks safety

arXiv AI

Sep 15

LLM Probability Concentration: How Alignment Shrinks the Generative Horizon

arXiv:2506.17871v4 Announce Type: replace-cross Abstract: Despite their impressive capabilities, aligned large language models (LLMs) often generate outputs that lack diversity. What drives this cons...

By Chenghao Yang, Sida Li, Ari Holtzman

llms safety

arXiv Machine Learning

Aug 5

Omega-S: A Functional Resilience Index for LLM Fine-Tuning

arXiv:2608. 03887v1 Announce Type: new Abstract: Fine-tuning a large language model on new data degrades what it previously learned.

By Alberto Acedo

llms fine-tuning benchmarks safety

One email a morning, machine-written

Related stories

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

Evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment

Constitutional Midtraining: Content Presence Drives Alignment Gains

Similar Models Learn Differently: Final-Window Pretraining Shapes Post-Training Beyond SFT

LLM Probability Concentration: How Alignment Shrinks the Generative Horizon

Omega-S: A Functional Resilience Index for LLM Fine-Tuning