Results 1 to 10 of about 631,318 (286)
Word Representation Learning in Multimodal Pre-Trained Transformers: An Intrinsic Evaluation
Transactions of the Association for Computational Linguistics, 2021 This study carries out a systematic intrinsic evaluation of the semantic representations learned by state-of-the-art pre-trained multimodal Transformers.
Sandro Pezzelle +2 more
doaj +1 more source
On the Solvability of Inductive Problems: A Study in Epistemic Topology [PDF]
Electronic Proceedings in Theoretical Computer Science, 2016 We investigate the issues of inductive problem-solving and learning by doxastic agents. We provide topological characterizations of solvability and learnability, and we use them to prove that AGM-style belief revision is "universal", i.e., that every ...
Alexandru Baltag +2 more
doaj +1 more source
Neural tracking of phrases in spoken language comprehension is automatic and task-dependent
eLife, 2022 Linguistic phrases are tracked in sentences even though there is no one-to-one acoustic phrase marker in the physical signal. This phenomenon suggests an automatic tracking of abstract linguistic structure that is endogenously generated by the brain ...
Sanne ten Oever +3 more
doaj +1 more source
Transactions of the Association for Computational Linguistics, 2022 We investigate the extent to which modern neural language models are susceptible to structural priming, the phenomenon whereby the structure of a sentence makes the same structure more probable in a follow-up sentence.
Arabella Sinclair +3 more
doaj +1 more source
Reasoning about Quantum Information: An Overview of Quantum Dynamic Logic
Applied Sciences, 2022 This paper provides an overview of quantum dynamic logics, showing how they have been designed and illustrating how these logics can be applied to verify the correctness of quantum protocols.
Alexandru Baltag, Sonja Smets
doaj +1 more source
Erasure of Unaligned Attributes from Neural Representations
Transactions of the Association for Computational Linguistics, 2023 We present the Assignment-Maximization Spectral Attribute removaL (AMSAL) algorithm, which erases information from neural representations when the information to be erased is implicit rather than directly being aligned to each input example.
Shun Shao, Yftah Ziser, Shay B. Cohen
doaj +1 more source
SAGE Open, 2023 As an emerging business model, live broadcast e-commerce is growing rapidly in China in recent years and prompts the emergence of network anchors, attracting the customers to buy a variety of goods in their live broadcasts.
Hongzheng Li, Ruojin Wang, Can Shi
doaj +1 more source
Monotonicity in a Numeral Classifier language
Glossa, 2023 While many languages require an obligatory plural morpheme to make reference to plural individuals, numeral classifier languages generally do not (Greenberg 1972; Sanches & Slobin 1973; Doetjes 2012: a.o.).
Takanobu Nakamura
doaj +2 more sources
Exhaustification in the semantics of cause and because
Glossa, 2023 We show that a single operation can account for three seemingly distinct properties of the semantics of cause and because. The properties are, firstly, their comparative nature: interpreting cause and because involves comparing what would happen in the ...
Dean Michael McHugh
doaj +2 more sources
Transactions of the Association for Computational Linguistics, 2021 We present a memory-based model for context- dependent semantic parsing. Previous approaches focus on enabling the decoder to copy or modify the parse from the previous utterance, assuming there is a dependency between the current and previous parses. In
Parag Jain, Mirella Lapata
doaj +1 more source