Zobrazeno 1 - 10
of 971
pro vyhledávání: '"Bohannon, John"'
A dense passage retrieval system can serve as the initial stages of information retrieval, selecting the most relevant text passages for downstream tasks. In this work we conducted experiments with the goal of finding how much the quality of a multil
Externí odkaz:
http://arxiv.org/abs/2407.00923
Autor:
Vasilyev, Oleg, Bohannon, John
We found that a simple property of clusters in a clustered dataset of news correlate strongly with importance and urgency of news (IUN) as assessed by LLM. We verified our finding across different news datasets, dataset sizes, clustering algorithms a
Externí odkaz:
http://arxiv.org/abs/2402.10302
Semantics of a sentence is defined with much less ambiguity than semantics of a single word, and we assume that it should be better preserved by translation to another language. If multilingual sentence embeddings intend to represent sentence semanti
Externí odkaz:
http://arxiv.org/abs/2305.14256
Autor:
Vasilyev, Oleg, Bohannon, John
We propose a new kind of embedding for natural language text that deeply represents semantic meaning. Standard text embeddings use the outputs from hidden layers of a pretrained language model. In our method, we let a language model learn from the te
Externí odkaz:
http://arxiv.org/abs/2208.08386
Transformer language models provide superior accuracy over previous models but they are computationally and environmentally expensive. Borrowing the concept of model cascading from computer vision, we introduce BabyBear, a framework for cascading mod
Externí odkaz:
http://arxiv.org/abs/2205.11747
We propose a simple and practical method for named entity linking (NEL), based on entity representation by multiple embeddings. To explore this method, and to review its dependency on parameters, we measure its performance on Namesakes, a highly chal
Externí odkaz:
http://arxiv.org/abs/2205.10498
Autor:
Vasilyev, Oleg, Bohannon, John
Factual consistency is one of important summary evaluation dimensions, especially as summary generation becomes more fluent and coherent. The ESTIME measure, recently proposed specifically for factual consistency, achieves high correlations with huma
Externí odkaz:
http://arxiv.org/abs/2112.11638
We present Namesakes, a dataset of ambiguously named entities obtained from English-language Wikipedia and news articles. It consists of 58862 mentions of 4148 unique entities and their namesakes: 1000 mentions from news, 28843 from Wikipedia article
Externí odkaz:
http://arxiv.org/abs/2111.11372
The creation of a quality summarization dataset is an expensive, time-consuming effort, requiring the production and evaluation of summaries by both trained humans and machines. If such effort is made in one language, it would be beneficial to be abl
Externí odkaz:
http://arxiv.org/abs/2109.08129
Evaluating large summarization corpora using humans has proven to be expensive from both the organizational and the financial perspective. Therefore, many automatic evaluation metrics have been developed to measure the summarization quality in a fast
Externí odkaz:
http://arxiv.org/abs/2105.06027