Zobrazeno 1 - 10 of 205 pro vyhledávání: '"Sadao Kurohashi"'
2
All-in-One: Emotion, Sentiment and Intensity Prediction Using a Multi-Task Ensemble Framework
Autor: Deepanway Ghosal, Pushpak Bhattacharyya, Shad Akhtar, Asif Ekbal, Sadao Kurohashi
Publikováno v: IEEE Transactions on Affective Computing. 13:285-297
We propose a multi-task ensemble framework that jointly learns multiple related problems. The ensemble model aims to leverage the learned representations of three deep learning models (i.e., CNN, LSTM and GRU) and a hand-crafted feature representatio
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::35c9d16dcc19eade11b32b4414e52680
https://doi.org/10.1109/taffc.2019.2926724
Zobrazit plný text záznamu
3
RODA: Reverse Operation Based Data Augmentation for Solving Math Word Problems
Autor: Daisuke Kawahara, Fei Cheng, Qianying Liu, Sujian Li, Sadao Kurohashi, Wenyu Guan
Publikováno v: IEEE/ACM Transactions on Audio, Speech, and Language Processing. 30:1-11
Automatically solving math word problems is a critical task in the field of natural language processing. Recent models have reached their performance bottleneck and require more high-quality data for training. We propose a novel data augmentation met
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::8c4ce9b50349fa08cce472c58d6e36d8
https://doi.org/10.1109/taslp.2021.3126932
Zobrazit plný text záznamu
4
A Peek Into the Memory of T5: Investigating the Factual Knowledge Memory in a Closed-Book QA Setting and Finding Responsible Parts
Autor: Tareq Alkhaldi, Chenhui Chu, Sadao Kurohashi
Publikováno v: Journal of Natural Language Processing. 29:762-784
Recent research shows that Transformer-based language models (LMs) store considerable factual knowledge from the unstructured text datasets on which they are pre-trained. The existence and amount of such knowledge have been investigated by probing pr
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3f749819f1bf629cf94eaf40f3fe7e0a
https://doi.org/10.5715/jnlp.29.762
Zobrazit plný text záznamu
9
Flexibly Focusing on Supporting Facts, Using Bridge Links, and Jointly Training Specialized Modules for Multi-Hop Question Answering
Autor: Sadao Kurohashi, Chenhui Chu, Tareq Alkhaldi
Publikováno v: IEEE/ACM Transactions on Audio, Speech, and Language Processing. 29:3216-3225
With the help of the detailed annotated question answering dataset HotpotQA, recent question answering models are trained to justify their predicted answers with supporting facts from context documents. Some related works train the same model to find
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::48c51241d8d961dc4dfc5fd48408354d
https://doi.org/10.1109/taslp.2021.3120643
Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání

Omezení vyhledávání
Plný text Recenzováno Digitální knihovna AV ČR
Zdroje