Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Ferritto, Anthony"'
Autor:
Deshpande, Ameet, Sultan, Md Arafat, Ferritto, Anthony, Kalyan, Ashwin, Narasimhan, Karthik, Sil, Avirup
Fine-tuning pre-trained language models (PLMs) achieves impressive performance on a range of downstream tasks, and their sizes have consequently been getting bigger. Since a different copy of the model is required for each task, this paradigm is infe
Externí odkaz:
http://arxiv.org/abs/2211.16634
Autor:
McCarley, Scott, Bornea, Mihaela, Rosenthal, Sara, Ferritto, Anthony, Sultan, Md Arafat, Sil, Avirup, Florian, Radu
Recent machine reading comprehension datasets include extractive and boolean questions but current approaches do not offer integrated support for answering both question types. We present a multilingual machine reading comprehension system and front-
Externí odkaz:
http://arxiv.org/abs/2206.08441
Existing models on Machine Reading Comprehension (MRC) require complex model architecture for effectively modeling long texts with paragraph representation and classification, thereby making inference computationally inefficient for production use. I
Externí odkaz:
http://arxiv.org/abs/2105.03229
Autor:
Zhang, Rong, Reddy, Revanth Gangi, Sultan, Md Arafat, Castelli, Vittorio, Ferritto, Anthony, Florian, Radu, Kayi, Efsun Sarioglu, Roukos, Salim, Sil, Avirup, Ward, Todd
Transfer learning techniques are particularly useful in NLP tasks where a sizable amount of high-quality annotated data is difficult to obtain. Current approaches directly adapt a pre-trained language model (LM) on in-domain text before fine-tuning t
Externí odkaz:
http://arxiv.org/abs/2010.05904
Autor:
Castelli, Vittorio, Chakravarti, Rishav, Dana, Saswati, Ferritto, Anthony, Florian, Radu, Franz, Martin, Garg, Dinesh, Khandelwal, Dinesh, McCarley, Scott, McCawley, Mike, Nasr, Mohamed, Pan, Lin, Pendus, Cezar, Pitrelli, John, Pujar, Saurabh, Roukos, Salim, Sakrajda, Andrzej, Sil, Avirup, Uceda-Sosa, Rosario, Ward, Todd, Zhang, Rong
We introduce TechQA, a domain-adaptation question answering dataset for the technical support domain. The TechQA corpus highlights two real-world issues from the automated customer support domain. First, it contains actual questions posed by users on
Externí odkaz:
http://arxiv.org/abs/1911.02984
Autor:
Ferritto, Anthony, Pan, Lin, Chakravarti, Rishav, Roukos, Salim, Florian, Radu, Murdock, J. William, Sil, Avirup
Many of the top question answering systems today utilize ensembling to improve their performance on tasks such as the Stanford Question Answering Dataset (SQuAD) and Natural Questions (NQ) challenges. Unfortunately most of these systems do not publis
Externí odkaz:
http://arxiv.org/abs/1911.00337
Autor:
Pan, Lin, Chakravarti, Rishav, Ferritto, Anthony, Glass, Michael, Gliozzo, Alfio, Roukos, Salim, Florian, Radu, Sil, Avirup
Existing literature on Question Answering (QA) mostly focuses on algorithmic novelty, data augmentation, or increasingly large pre-trained language models like XLNet and RoBERTa. Additionally, a lot of systems on the QA leaderboards do not have assoc
Externí odkaz:
http://arxiv.org/abs/1909.05286
Autor:
Glass, Michael, Gliozzo, Alfio, Chakravarti, Rishav, Ferritto, Anthony, Pan, Lin, Bhargav, G P Shrivatsa, Garg, Dinesh, Sil, Avirup
BERT (Bidirectional Encoder Representations from Transformers) and related pre-trained Transformers have provided large gains across many language understanding tasks, achieving a new state-of-the-art (SOTA). BERT is pre-trained on two auxiliary task
Externí odkaz:
http://arxiv.org/abs/1909.04120
Autor:
Chakravarti, Rishav, Pendus, Cezar, Sakrajda, Andrzej, Ferritto, Anthony, Pan, Lin, Glass, Michael, Castelli, Vittorio, Murdock, J. William, Florian, Radu, Roukos, Salim, Sil, Avirup
Publikováno v:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations
This paper introduces a novel orchestration framework, called CFO (COMPUTATION FLOW ORCHESTRATOR), for building, experimenting with, and deploying interactive NLP (Natural Language Processing) and IR (Information Retrieval) systems to production envi
Externí odkaz:
http://arxiv.org/abs/1908.06121