Papers by Nafis Sadeq
Unsupervised Improvement of Factual Knowledge in Language Models (2023.eacl-main)
Copied to clipboard
| Challenge: | Masked language modeling (MLM) is often dominated by high-frequency words that are sub-optimal for learning factual knowledge. |
| Approach: | They propose an approach that forces the model to prioritize informative words in a fully unsupervised way. |
| Outcome: | The proposed approach significantly improves the performance of pretrained language models on factual recall, question answering, sentiment analysis, and natural language inference in a closed-book setting. |
InforMask: Unsupervised Informative Masking for Language Model Pretraining (2022.emnlp-main)
Copied to clipboard
| Challenge: | Masked language modeling is used for pretraining large language models for knowledge-intensive tasks. |
| Approach: | They propose an unsupervised masking strategy that exploits Pointwise Mutual Information to select the most informative tokens to mask. |
| Outcome: | The proposed strategy outperforms random masking and previously proposed masking strategies on the factual recall benchmark LAMA and the question answering benchmark SQuAD v1 and v2. |
Improving End-to-End Bangla Speech Recognition with Semi-supervised Training (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to train speech recognition systems require large annotated corpus. |
| Approach: | They propose a semi-supervised training approach that exploits large unpaired audio and text data to improve the performance of an automatic speech recognition system. |
| Outcome: | The proposed method reduces the WER of the system from 37% to 31.9%. |
Preparation of Bangla Speech Corpus from Publicly Available Audio & Text (2020.lrec-1)
Copied to clipboard
Shafayat Ahmed, Nafis Sadeq, Sudipta Saha Shubha, Md. Nahidul Islam, Muhammad Abdullah Adnan, Mohammad Zuberul Islam
| Challenge: | Automated speech recognition systems require large annotated speech corpus for training. |
| Approach: | They propose to use publicly available Bangla audiobooks and TV news recordings as input to prepare a large speech corpus with reasonable confidence. |
| Outcome: | The proposed algorithm outperforms the existing speech corpus and the existing corpus with speaker diarization and gender detection. |
Customizing Grapheme-to-Phoneme System for Non-Trivial Transcription Problems in Bangla Language (N19-1)
Copied to clipboard
Sudipta Saha Shubha, Nafis Sadeq, Shafayat Ahmed, Md. Nahidul Islam, Muhammad Abdullah Adnan, Md. Yasin Ali Khan, Mohammad Zuberul Islam
| Challenge: | Existing methods for Grapheme to phoneme conversion in Bangla language are mostly rule-based. |
| Approach: | They propose to use a lexicon to train a robust Grapheme to phoneme conversion system in Bangla language. |
| Outcome: | The proposed method outperforms other state-of-the-art approaches for G2P conversion in Bangla language. |
Mitigating Hallucination in Fictional Character Role-Play (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Influence of parametric knowledge of large language models (LLMs) often causes role-playing characters to act out of character and hallucinate about things outside the scope of their knowledge. |
| Approach: | They propose a method that modulates the influence of parametric knowledge using a pre-calibrated confidence threshold to mitigate hallucination in fictional character role-play. |
| Outcome: | The proposed method reduces the factual accuracy of generated responses by 18% for adversarial questions and 44% in temporal hallucination for time-sensitive interviews. |