Results 1 to 10 of about 1,443,727 (379)
SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension [PDF]
arXiv.org, 2023 Based on powerful Large Language Models (LLMs), recent generative Multimodal Large Language Models (MLLMs) have gained prominence as a pivotal research area, exhibiting remarkable capability for both comprehension and generation. In this work, we address
Bohao Li +5 more
semanticscholar +1 more source
DreamLLM: Synergistic Multimodal Comprehension and Creation [PDF]
International Conference on Learning Representations, 2023 This paper presents DreamLLM, a learning framework that first achieves versatile Multimodal Large Language Models (MLLMs) empowered with frequently overlooked synergy between multimodal comprehension and creation.
Runpei Dong +13 more
semanticscholar +1 more source
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension [PDF]
Annual Meeting of the Association for Computational Linguistics, 2019 We present BART, a denoising autoencoder for pretraining sequence-to-sequence models. BART is trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
M. Lewis +7 more
semanticscholar +1 more source
Comprehensive comprehensions [PDF]
Proceedings of the ACM SIGPLAN workshop on Haskell workshop, 2007 We propose an extension to list comprehensions that makes it easy to express the kind of queries one would write in SQL using ORDER BY, GROUP BY, and LIMIT. Our extension adds expressive power to comprehensions, and generalises the SQL constructs that inspired it. It is easy to implement, using simple desugaring rules.
Simon Jones, Philip Wadler
openaire +2 more sources
SQuAD: 100,000+ Questions for Machine Comprehension of Text [PDF]
Conference on Empirical Methods in Natural Language Processing, 2016 We present the Stanford Question Answering Dataset (SQuAD), a new reading comprehension dataset consisting of 100,000+ questions posed by crowdworkers on a set of Wikipedia articles, where the answer to each question is a segment of text from the ...
Pranav Rajpurkar +3 more
semanticscholar +1 more source
High tibial osteotomy: A review of the readability and quality of patientinformation on the internet [PDF]
Health Promotion Perspectives, 2021 Background: High tibial osteotomy (HTO) is a common procedure performed for unicompartmental knee osteoarthritis (OA). Patients are increasingly using the internet to research surgical procedures to help aid decision making.
Matthew Clark +4 more
doaj +1 more source
TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension [PDF]
Annual Meeting of the Association for Computational Linguistics, 2017 We present TriviaQA, a challenging reading comprehension dataset containing over 650K question-answer-evidence triples. TriviaQA includes 95K question-answer pairs authored by trivia enthusiasts and independently gathered evidence documents, six per ...
Mandar Joshi +3 more
semanticscholar +1 more source
But is it Comprehensive? Unpacking the ‘comprehensive’ in comprehensive sexuality education [PDF]
Health Education Journal, 2020 Background: Comprehensive Sexuality Education (CSE) is increasingly gaining traction within the international community. CSE is regarded as an important means of informing young people about their rights and sexual health, improving public health outcomes and contributing to sustainable development.
Esther Miedema +2 more
openaire +3 more sources
Is the Comprehensive Assessment Really Comprehensive? [PDF]
SSRN Electronic Journal, 2014 Analyzing the database made available by the European Central Bank (ECB) we evaluate the Comprehensive Assessment (asset quality review and stress test). We provide some insights on the supervisory approach of the ECB and on banking regulation. In a nutshell the main results are: i) risk adjusted capital ratios are negatively related to the asset ...
BARUCCI, EMILIO +2 more
openaire +3 more sources
RACE: Large-scale ReAding Comprehension Dataset From Examinations [PDF]
Conference on Empirical Methods in Natural Language Processing, 2017 We present RACE, a new dataset for benchmark evaluation of methods in the reading comprehension task. Collected from the English exams for middle and high school Chinese students in the age range between 12 to 18, RACE consists of near 28,000 passages ...
Guokun Lai +4 more
semanticscholar +1 more source