paper-with-me

Papers

Emergent Alignment

2026-06-17 · Martin Kolář arxiv

Can Large Language Models (LLMs) discern when their own outputs are misaligned with human ethics? And can they self-correct? We endow an LLM with a conscience step that reviews its own reasoning and outputs, and we extend the training loss with an alignment component using Direct Preference Optimization (DPO) to steer the model away from non-ethical outputs. The result is an online technique to align models in a wide range of applications: training, fine-tuning, adversarial prompting, and zero-shot learning. It does not require a weaker or stronger judge, relying instead on a frozen copy of itself. In previous work, the Emergent Misalignment scenario showed a range of emergent unethical behaviors from fine-tuning the model to hack code. Instead, we empirically show how to achieve Emergent Alignment: a single high-level introspective question steers training toward an ethical model under the same code hacking scenario.

📄 PDF Abstract BibTeX arXiv:2606.19527

Code (0)

등록된 구현이 없습니다.

Tasks

Zero-Shot Learning

Similar Papers 제목 키워드 기반

Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment

2026-02-16 · Laurène Vaugrante, Anietta Weckauff, Thilo Hagendorff arxiv

Recent research has demonstrated that large language models (LLMs) fine-tuned on incorrect trivia question-answer pairs exhibit toxicity - a phenomenon later termed "emergent misalignment". Moreover, research has shown t…

BLOCK-EM: Preventing Emergent Misalignment via Latent Blocking

2026-01-31 · Muhammed Ustaomeroglu, Guannan Qu arxiv

Emergent misalignment can arise when a language model is fine-tuned on a narrowly scoped supervised objective: the model learns the target behavior, yet also develops undesirable out-of-domain behaviors. We investigate a…

Emergent Misalignment Can Be Induced by Sycophancy and Reversed via Alignment Gating

2026-06-08 · Sicheng Wang, Xiangyang Zhu, Han Wang, Zongrui Wang 외 arxiv

Prior work has shown that fine-tuning large language models on malicious or incorrect outputs in narrow domains can induce broad misalignment and harmful behavior, a phenomenon known as emergent misalignment. However, ef…

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

2025-02-24 · Jan Betley, Daniel Tan, Niels Warncke, Anna Sztyber-Betley 외

We present a surprising result regarding LLMs and alignment. In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting model acts misaligned on a broad range of pr…

Computer Security

The Devil in the Details: Emergent Misalignment, Format and Coherence in Open-Weights LLMs

2025-11-25 · Craig Dickson arxiv

Prior work has shown that fine-tuning models on a narrow domain with misaligned data can lead to broad misalignment - a phenomenon termed "emergent misalignment" (Betley et al. 2025). While all tested models were suscept…

Code Generation