Results 61 to 70 of about 14,332 (195)

Tibetan Data Augmentation via GAN‐Based Handwritten Text Generation

CAAI Transactions on Intelligence Technology, Volume 11, Issue 1, Page 55-65, February 2026.
ABSTRACT Increased awareness of Tibetan cultural preservation, along with technological advancements, has led to significant efforts in academic research on Tibetan. However, the structural complexity of the Tibetan language and limited labeled handwriting data impede advancements in Optical Character Recognition (OCR) and other applications.
Dorje Tashi, Bingtian Chen, Tianying Sheng, Yongbin Yu, Xiangxiang Wang, Jin Zhang, Lobsang Yeshi, Rinchen Dongrub, Thupten Tsering, Nyima Tashi   +9 more
wiley   +1 more source

Which Words Are Special? Identification of “Sight” Words in Educational Resources

Reading Research Quarterly, Volume 61, Issue 1, January/February/March 2026.
We examined the overlap in words and properties in the contents of six commonly used word lists designed to support early word reading skills. The resources overlapped very little in their contents, including dramatic variation in terms of the properties of the words included in each.
Matthew J. Cooper Borkenhagen, Lauren P. Schilling, Mark S. Seidenberg   +2 more
wiley   +1 more source

BERT-NAR-BERT: A Non-Autoregressive Pre-Trained Sequence-to-Sequence Model Leveraging BERT Checkpoints

IEEE Access
We introduce BERT-NAR-BERT (BnB) – a pre-trained non-autoregressive sequence-to-sequence model, which employs BERT as the backbone for the encoder and decoder for natural language understanding and generation tasks.
Mohammad Golam Sohrab, Masaki Asada, Matiss Rikters, Makoto Miwa   +3 more
doaj   +1 more source

Multilingual Adaptation of RNN Based ASR Systems

, 2018
In this work, we focus on multilingual systems based on recurrent neural networks (RNNs), trained using the Connectionist Temporal Classification (CTC) loss function. Using a multilingual set of acoustic units poses difficulties.
Müller, Markus, Stüker, Sebastian, Waibel, Alex   +2 more
core   +1 more source

The Agent Preference in Ontogeny: Predictability of Agent and Patient Roles in Child‐Directed Utterances Across Languages

Cognitive Science, Volume 50, Issue 1, January 2026.
Abstract Language comprehension unfolds incrementally, requiring listeners to continually predict and revise interpretations. Comprehenders across very diverse languages show a consistent preference for agents, anticipating the agent (“the doer” of an action) more strongly than the patient (“the undergoer”). An unresolved question is how the preference
Eva Huber, Aylin C. Küntay, Balthasar Bickel, Sabine Stoll   +3 more
wiley   +1 more source

Shared Minds: The Cognitive Parallels Between Humans and Artificial Intelligence

Human Behavior and Emerging Technologies, Volume 2026, Issue 1, 2026.
This narrative review integrates evidence from cognitive science and AI research to challenge commonly accepted dichotomies between human and artificial cognition, such as the assumed divide between genuine human understanding and mere machine pattern matching. Instead, we propose a view that recognises similarities in their cognitive architectures and
Sébastien Tremblay, Alexandre Marois, Marzieh Zare, Daniel Lafond, Tze Wei Liew   +4 more
wiley   +1 more source

Lifelong Neural Predictive Coding: Learning Cumulatively Online without Forgetting

, 2021
In lifelong learning systems, especially those based on artificial neural networks, one of the biggest obstacles is the severe inability to retain old knowledge as new information is encountered. This phenomenon is known as catastrophic forgetting.
Giles, C. Lee, Kifer, Daniel, Mali, Ankur, Ororbia, Alexander   +3 more
core  

In Situ Graph Reasoning and Knowledge Expansion Using Graph‐PRefLexOR

Advanced Intelligent Discovery, Volume 1, Issue 3, December 2025.
Graph‐PRefLexOR is a novel framework that enhances language models with in situ graph reasoning, symbolic abstraction, and recursive refinement. By integrating graph‐based representations into generative tasks, the approach enables interpretable, multistep reasoning.
Markus J. Buehler
wiley   +1 more source

LSE-FS-UVigo Dataset and Keypoint-Based Fingerspelling Recognition

IEEE Access
Continuous fingerspelling recognition remains a challenging task due to rapid hand and finger movements, coarticulation, and the limited availability of annotated data across sign languages. We introduce LSE-FS-UVigo, the first open dataset of continuous
Jose Luis Ruanova-Lea, Jose Luis Alba-Castro, Laura Docio-Fernandez   +2 more
doaj   +1 more source

Connectionist Temporal Modeling for Weakly Supervised Action Labeling

, 2016
We propose a weakly-supervised framework for action labeling in video, where only the order of occurring actions is required during training time. The key challenge is that the per-frame alignments between the input (video) and label (action) sequences ...
A Graves, Chih-Chung Chang, DE Rumelhart, GW Taylor, JC Niebles, P Bojanowski, R Achanta   +6 more
core   +1 more source