arXiv:2608. 04347v1 Announce Type: new Abstract: Fine-tuning enables a source model to acquire desired capabilities and behaviors in a target domain while retaining much of its general-purpose competence.
By Kotaro Yoshida, Laura Gomezjurado Gonzalez, Yukinori Yamamoto, Yuji Naraki, Ryotaro Shimizu, Wenya Wang
Fine-tuning enables a source model to acquire desired capabilities and behaviors in a target domain while retaining much of its general-purpose competence. However, this adaptation process can also degrade alignment properties that were present in the source model.
arXiv:2606. 03810v1 Announce Type: cross Abstract: Consistency training encourages a model to produce similar outputs across related inputs or sampling procedures.
By David Demitri Africa, Arathi Mani
The paper "Stress-testing Alignment Midtraining" examines the effectiveness of alignment midtraining (AMT), a technique that continues pretraining on alignment-relevant data to improve generalisation beyond post‑training methods. Experiments on models up to 110 billion parameters and 1 billion midtraining tokens reveal that AMT can steer a model’s motivation in simple scenarios, but its effects are quickly overridden by even a tiny fraction of finetuning data with a competing motivation. The study also shows that rule-following requires demonstrations in either the midtraining or post‑training datasets to be robustly learned, leading the authors to conclude that current public evidence is insufficient to confirm that AMT resolves the core alignment challenges of powerful AI systems.
By Sid Baines, Jonathan Bostock, Maria Angelica Martinez, Andrew Draganov, David Africa, Daniel Tan
arXiv:2607. 18114v1 Announce Type: cross Abstract: Modern LLMs are alarmingly susceptible to surprisingly simple immaterial changes of input prompts: a casual hint, an incorrectly labeled few-shot example, or a fake prior assistant turn often flips an originally correct answer.
By Prakhar Gupta, Terry Jingchen Zhang, Florent Draye, Bernhard Sch\"olkopf, Zhijing Jin
The paper investigates whether automated alignment researchers (AARs) can post‑train language models to reduce well‑characterized alignment failures such as deception, sycophancy, and jailbreaks while preserving general capability. Across ten failures, the strongest AAR methods significantly lower targeted failures and generalize to held‑out benchmarks, larger models, and multi‑turn audits. In contrast, a human baseline of 28 experienced researchers, given eight hours to devise one‑shot methods, underperformed the best AAR approaches, and providing human ideas to AARs did not improve results.
By Chen Yueh-Han, Jiaxin Wen, Jan Hendrik Kirchner