This work proposes a flexible architecture to integrate text- and image-based distributional information, and shows in a set of empirical tests that the integrated model is superior to the purely text-based approach, and it provides somewhat complementary semantic information with respect to the latter.
While visual models with state-of-the-art computer vision techniques perform worse than textual models in general tasks, they are as good or better models of the meaning of words with visual correlates such as color terms, even in a nontrivial task that involves nonliteral uses of such words.
This paper proposes an optimization framework and demonstrates the use of Learning To Rank (LTR) to automatically construct timeline summaries from Web news articles, showing that this approach outperforms existing solutions in producing high quality Timeline summaries.
This paper proposes Multihop Attention Networks (MAN) which use multiple vectors which focus on different parts of the question for its overall semantic representation and apply multiple steps of attention to learn representations for the candidate answers.
This paper proposes an entity-oriented search system to support retrieval and analytics on the Internet Archive that complements existing web archive search tools through a user-friendly interface, and provides a great benefit of taking user feedback on the current web into account also forweb archive search.
This work proposes a novel model that complements traditional text-based approaches by rewarding entities that exhibit a high temporal correlation with topics during their burst time period by exploiting temporal information from the Wikipedia edit history and page view logs.
Experimental results show that the proposed approaches retrieve contextualization information for older articles from the New York Times Archive with high precision and outperform baselines significantly.
This paper proposes a novel method of computing the contextual relatedness of entity relatedness with integrated time and topic models, and shows that the proposed relatedness can effectively recommend entities, taking contexts into account.
The results show that the proposed methods for tackling the problem of cross-device matching for online advertising at CIKM Cup 2016 obtain promising results, in which the ranking-based method outperforms the classification- based method for the task.
A pipeline system and preliminary results for Tweet Contextualization at INEX 2013 are described and it is shown that the approach does not work well where the run was ranked 22nd out of 24 runs.