Picture for Md Arafat Sultan

Md Arafat Sultan

IBM Research AI, T.J. Watson Research Center, New York, USA

Synthetic Target Domain Supervision for Open Retrieval QA

Apr 20, 2022
Revanth Gangi Reddy, Bhavani Iyer, Md Arafat Sultan, Rong Zhang, Avirup Sil, Vittorio Castelli, Radu Florian, Salim Roukos

Figure 1 for Synthetic Target Domain Supervision for Open Retrieval QA

Figure 2 for Synthetic Target Domain Supervision for Open Retrieval QA

Figure 3 for Synthetic Target Domain Supervision for Open Retrieval QA

Figure 4 for Synthetic Target Domain Supervision for Open Retrieval QA

* Published at SIGIR 2021

Access Paper or Ask Questions

Answer Span Correction in Machine Reading Comprehension

Nov 06, 2020
Revanth Gangi Reddy, Md Arafat Sultan, Efsun Sarioglu Kayi, Rong Zhang, Vittorio Castelli, Avirup Sil

Figure 1 for Answer Span Correction in Machine Reading Comprehension

Figure 2 for Answer Span Correction in Machine Reading Comprehension

Figure 3 for Answer Span Correction in Machine Reading Comprehension

Figure 4 for Answer Span Correction in Machine Reading Comprehension

* Accepted in Findings of EMNLP 2020

Access Paper or Ask Questions

Improved Synthetic Training for Reading Comprehension

Oct 24, 2020
Yanda Chen, Md Arafat Sultan, Vittorio Castelli

Figure 1 for Improved Synthetic Training for Reading Comprehension

Figure 2 for Improved Synthetic Training for Reading Comprehension

Figure 3 for Improved Synthetic Training for Reading Comprehension

Figure 4 for Improved Synthetic Training for Reading Comprehension

* 11 pages, 2 figures

Access Paper or Ask Questions

Multi-Stage Pre-training for Low-Resource Domain Adaptation

Oct 12, 2020
Rong Zhang, Revanth Gangi Reddy, Md Arafat Sultan, Vittorio Castelli, Anthony Ferritto, Radu Florian, Efsun Sarioglu Kayi, Salim Roukos, Avirup Sil, Todd Ward

Figure 1 for Multi-Stage Pre-training for Low-Resource Domain Adaptation

Figure 2 for Multi-Stage Pre-training for Low-Resource Domain Adaptation

Figure 3 for Multi-Stage Pre-training for Low-Resource Domain Adaptation

Figure 4 for Multi-Stage Pre-training for Low-Resource Domain Adaptation

* Accepted at EMNLP 2020

Access Paper or Ask Questions

GPT-too: A language-model-first approach for AMR-to-text generation

May 27, 2020
Manuel Mager, Ramon Fernandez Astudillo, Tahira Naseem, Md Arafat Sultan, Young-Suk Lee, Radu Florian, Salim Roukos

Figure 1 for GPT-too: A language-model-first approach for AMR-to-text generation

Figure 2 for GPT-too: A language-model-first approach for AMR-to-text generation

Figure 3 for GPT-too: A language-model-first approach for AMR-to-text generation

Figure 4 for GPT-too: A language-model-first approach for AMR-to-text generation

* Paper accepted to the Annual Meeting of the Association for Computational Linguistics (ACL 2020)

Access Paper or Ask Questions