Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Florian Mai
Computer Science > Computation and Language
Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A leading mechanistic account attributes EM to persona features: latent directions acquired during pre-training that misaligned fine-tuning amplifies.
Computer Science > Computation and Language
Large language models (LLMs) have developed rapidly and become valuable tools in everyday life. However, how to align LLMs to a particular set of human values is still an open problem. Recent studies show that instruction tuning has strong potential for zero-shot tasks and may serve as an effective approach to addressing value alignment.
Computer Science > Artificial Intelligence
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Computer Science > Computation and Language
Large language models can hide hidden behaviors that activate only under narrow conditions, such as backdoor triggers, sleeper-agent deployment cues, sandbagging, or topic-conditioned censorship. Such behaviors are difficult to detect without prior knowledge what to look for. We present activation-matched finetuning, an unsupervised detection method that assumes no knowledge of the trigger or the target behavior.
Computer Science > Computation and Language
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
IKnow: Instruction-Knowledge-Aware Continual Pretraining for Effective Domain Adaptation
arXiv:2510.20377 (cs) View PDF HTML (experimental) Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2510.20377 [cs.AI] (or arXiv:2510.20377v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2510.20377 Submission history From: Tianyi Zhang [ view email] [v1] Thu, 23 Oct 2025 09:21:13 UTC (1,009 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Connected Papers Toggle Connected Papers (What is Connected Papers?)...
CO2 Conversion to Higher Hydrocarbons on Iron‐Based Catalysts: Promoter Impact and Kinetic Studies
1 Introduction A major challenge to mitigate the effects of climate change is the sustainable production of liquid fuels from renewable resources to decrease anthropogenic CO2 emissions. One option is the use of CO2 from various sources (power plants, chemical industry, and possibly separated from air) and subsequent conversion with renewable hydrogen to higher hydrocarbons (HCs).
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Florian
Contact Florian, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack