2 citations · 5 across the 10 of their papers we have counts for
5 papers · 1 filter
<think> So let's replace this phrase with insult... </think> Lessons learned from generation of toxic texts with LLMs
Sergey Pletenev, Daniil Moskovskiy, Alexander Panchenko
Modern Large Language Models (LLMs) are excellent at generating synthetic data. However, their performance in sensitive domains such as text detoxification has not received proper…
HatePRISM: Policies, Platforms, and Research Integration. Advancing NLP for Hate Speech Proactive Mitigation
Naquee Rizwan, Seid Muhie Yimam, Daryna Dementieva +11
Despite regulations imposed by nations and social media platforms, e.g. (Government of India, 2021; European Parliament and Council of the European Union, 2022), inter alia, hatefu…
Generalized Fisher-Weighted SVD: Scalable Kronecker-Factored Fisher Approximation for Compressing Large Language Models
Viktoriia Chekalina, Daniil Moskovskiy, Tatiana Matveeva +2
The Fisher information is a fundamental concept for characterizing the sensitivity of parameters in neural networks. However, leveraging the full observed Fisher information is too…
How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?
Sergey Pletenev, Maria Marina, Daniil Moskovskiy +4
The performance of Large Language Models (LLMs) on many tasks is greatly limited by the knowledge learned during pre-training and stored in the model's parameters. Low-rank adaptat…
SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators
Daniil Moskovskiy, Nikita Sushko, Sergey Pletenev +2
Existing approaches to multilingual text detoxification are hampered by the scarcity of parallel multilingual datasets. In this work, we introduce a pipeline for the generation of…