Zobrazeno 1 - 10
of 341
pro vyhledávání: '"Dimitrov Dimitar P"'
Autor:
Alexandrov, Anton, Raychev, Veselin, Dimitrov, Dimitar I., Zhang, Ce, Vechev, Martin, Toutanova, Kristina
We present BgGPT-Gemma-2-27B-Instruct and BgGPT-Gemma-2-9B-Instruct: continually pretrained and fine-tuned versions of Google's Gemma-2 models, specifically optimized for Bulgarian language understanding and generation. Leveraging Gemma-2's multiling
Externí odkaz:
http://arxiv.org/abs/2412.10893
Autor:
Beshirov, Angel, Dobreva, Milena, Dimitrov, Dimitar, Hardalov, Momchil, Koychev, Ivan, Nakov, Preslav
The digitization of historical documents is crucial for preserving the cultural heritage of the society. An important step in this process is converting scanned images to text using Optical Character Recognition (OCR), which can enable further search
Externí odkaz:
http://arxiv.org/abs/2409.00527
A key challenge of quantum programming is uncomputation: the reversible deallocation of qubits. And while there has been much recent progress on automating uncomputation, state-of-the-art methods are insufficient for handling today's expressive quant
Externí odkaz:
http://arxiv.org/abs/2406.14227
Federated learning works by aggregating locally computed gradients from multiple clients, thus enabling collaborative training without sharing private client data. However, prior work has shown that the data can actually be recovered by the server us
Externí odkaz:
http://arxiv.org/abs/2405.15586
Autor:
Linzbach, Stephan, Dimitrov, Dimitar, Kallmeyer, Laura, Evang, Kilian, Jabeen, Hajira, Dietze, Stefan
Pre-trained Language Models (PLMs) are known to contain various kinds of knowledge. One method to infer relational knowledge is through the use of cloze-style prompts, where a model is tasked to predict missing subjects or objects. Typically, designi
Externí odkaz:
http://arxiv.org/abs/2404.01992
Autor:
Das, Rocktim Jyoti, Hristov, Simeon Emilov, Li, Haonan, Dimitrov, Dimitar Iliyanov, Koychev, Ivan, Nakov, Preslav
We introduce EXAMS-V, a new challenging multi-discipline multimodal multilingual exam benchmark for evaluating vision language models. It consists of 20,932 multiple-choice questions across 20 school disciplines covering natural science, social scien
Externí odkaz:
http://arxiv.org/abs/2403.10378
Federated learning is a framework for collaborative machine learning where clients only share gradient updates and not their private data with a server. However, it was recently shown that gradient inversion attacks can reconstruct this data from the
Externí odkaz:
http://arxiv.org/abs/2403.03945
The wide-spread use of social networks has given rise to subjective, misleading, and even false information on the Internet. Thus, subjectivity detection can play an important role in ensuring the objectiveness and the quality of a piece of informati
Externí odkaz:
http://arxiv.org/abs/2309.06844
We study the behavior of the smallest possible constants $d(a,b)$ and $d_n$ in Hardy's inequalities $$ \int_a^b\left(\frac{1}{x}\int_a^xf(t)dt\right)^2\,dx\leq d(a,b)\,\int_a^b [f(x)]^2 dx $$ and $$ \sum_{k=1}^{n}\Big(\frac{1}{k}\sum_{j=1}^{k}a_j\Big
Externí odkaz:
http://arxiv.org/abs/2306.08172
Malicious server (MS) attacks have enabled the scaling of data stealing in federated learning to large batch sizes and secure aggregation, settings previously considered private. However, many concerns regarding the client-side detectability of MS at
Externí odkaz:
http://arxiv.org/abs/2306.03013