arXiv:2608. 16834v1 Announce Type: cross Abstract: We demonstrate that AI models are broadly susceptible to a phenomenon we call model hypnosis, in which individually weak and seemingly irrelevant cues in the prompt can be systematically combined to strongly control model behavior.
Subliminal Learning (SL) is a surprising type of generalization displayed by modern language models. It allows the transfer of a bias or behavior from a teacher model to a student by distilling from seemingly unrelated or random synthetic data from the teacher.
arXiv:2608. 05734v1 Announce Type: new Abstract: Subliminal Learning (SL) is a surprising type of generalization displayed by modern language models.
arXiv:2609.22119v1 Announce Type: cross Abstract: Evaluation awareness poses an unprecedented threat to model evaluation, but the mechanisms by which models detect it remain unknown. This study focus...
OpenAI’s new research explains why language models hallucinate. The findings show how improved evaluations can enhance AI reliability, honesty, and safety.
arXiv:2607. 29062v1 Announce Type: new Abstract: Model capabilities have improved in large part due to scaling chain of thought.
arXiv:2609.37568v1 Announce Type: new Abstract: Audio-visual large language models (AVLLMs) have made remarkable progress in multimodal understanding and reasoning through interactions among visual,...
Audio-visual large language models (AVLLMs) have made remarkable progress in multimodal understanding and reasoning through interactions among visual, auditory, and linguistic information. However, re...
The paper investigates why chain-of-thought (CoT) reasoning in language models is hard to control, attributing the issue to an entropy collapse in the first few tokens after the opening thinking tag. By applying two inference-time interventions—prefilling a single token and banning frequent opening tokens with in-context examples—the authors significantly improve CoT controllability, especially for addition and style tasks. The study also distinguishes between addition/style and suppression controllability, revealing different underlying mechanisms and correlations with monitorability across models.
The paper introduces SALVE, a method that uses text optimization to uncover and verbalize subliminal learning effects in language models. By optimizing a soft prompt and converting it into a legible text prompt, SALVE can reliably recover the teacher model’s hidden traits that are transmitted through a distillation dataset. The authors demonstrate SALVE’s effectiveness across various scenarios, including mixed datasets, biased teacher activation, and preference‑selected data, thereby providing a tool for detecting hidden influences in model training.
arXiv:2609.38962v1 Announce Type: new Abstract: Recent work on hallucination detection in large language models has shown that, for a fixed pre-trained model and reasoning task, it is possible to est...
arXiv:2606. 26874v1 Announce Type: new Abstract: Transcatheter Aortic Valve Replacement (TAVR) planning requires meticulous multimodal reasoning.
Whether large language models (LLMs) can control their own internal representations matters for both machine metacognition and AI safety. A recent study applied neurofeedback to LLMs and claimed that...