arXiv AI July 29, 2026 By Cen Lu, Yung-Chen Tang, Andrea Cavallaro

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.

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

Sep 10

The Geometry of Refusal: Why Post-Hoc Safety Is Fragile and Pretraining-Time Safety Persists

arXiv:2609.06934v1 Announce Type: cross Abstract: Post-hoc safety training (RLHF, DPO) is the dominant way to align large language models, yet jailbreaks (Zou et al., 2023b), fine-tuning attacks (Qi...

By Srikanth Malla, Chiho Choi, Joon Hee Choi

llms reinforcement-learning fine-tuning safety

arXiv AI

2d ago

Pre-training interventions, ex post facto: Grafting model beliefs across checkpoints

arXiv:2610.00767v1 Announce Type: cross Abstract: Pre-training interventions are critical to alignment research, since beliefs formed during pre-training shape how a model generalizes from later trai...

By Peter Nutter, Dani Roytburg, Cl\'ement Dumas, Jinghua Ou, Shi Feng

fine-tuning safety

arXiv Machine Learning

Aug 11

Early Data Exposure Improves Robustness to Subsequent Fine-Tuning

arXiv:2605. 12705v2 Announce Type: replace Abstract: How can we train models whose post-trained capabilities survive subsequent fine-tuning?

By Lawrence Feng, Gaurav R. Ghosal, Jacob Mitchell Springer, Ziqian Zhong, Aditi Raghunathan

fine-tuning

arXiv Machine Learning

Jul 30

Shared SFT Lessons Across Alignment, Model Organisms, and Toy Models

arXiv:2607. 26173v1 Announce Type: new Abstract: Alignment training, model organisms, and toy models are usually treated as separate research areas.

By Anton de la Fuente, Arthur Conmy

llms rag fine-tuning safety

arXiv Machine Learning

Jul 22

Mark, Don't Erase: Token Inoculation for Dual-Use Knowledge in LLMs

arXiv:2607. 18639v1 Announce Type: new Abstract: Safety interventions on dual-use knowledge typically choose between destroying hazardous content (e.

By Seunghyun Lee, Dongyoon Han, Sangdoo Yun

llms fine-tuning benchmarks 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

One email a morning, machine-written

Related stories

The Geometry of Refusal: Why Post-Hoc Safety Is Fragile and Pretraining-Time Safety Persists

Pre-training interventions, ex post facto: Grafting model beliefs across checkpoints

Early Data Exposure Improves Robustness to Subsequent Fine-Tuning

Shared SFT Lessons Across Alignment, Model Organisms, and Toy Models

Mark, Don't Erase: Token Inoculation for Dual-Use Knowledge in LLMs

Constitutional Midtraining: Content Presence Drives Alignment Gains