Papers by Ndapa Nakashole
Rethinking Self-Attention: Towards Interpretability in Neural Parsing (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work shows that attention mechanisms provide arguably explainable attention distributions that can help to interpret predictions. |
| Approach: | They propose a new self-attention layer where attention heads represent labels. |
| Outcome: | The proposed model obtains state-of-the-art results on the Penn Treebank and Chinese Treebank. |
Typology-Guided Adaptation in Multilingual Models (2025.acl-long)
Copied to clipboard
| Challenge: | Multilingual models often treat language diversity as a problem of data imbalance, overlooking structural variation. |
| Approach: | They propose a typologically grounded metric that quantifies how strongly a language relies on morphology for noun classification. |
| Outcome: | The proposed model outperforms baseline models on 10 Bantu languages . it improves Swahili accuracy by 14 points while maintaining performance on morphology-rich languages like Zulu . |
Medical Question Understanding and Answering with Knowledge Grounding and Semantic Self-Supervision (2022.coling-1)
Copied to clipboard
Khalil Mrini, Harpreet Singh, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter W. Chang, Emilia Farcas, Ndapa Nakashole
| Challenge: | Current medical question answering systems have difficulty processing long, detailed and informally worded questions . a growing number of approaches attempt to enhance the processing of consumer health questions - or medical question understanding . |
| Approach: | They propose a medical question understanding and answering system with knowledge grounding and semantic self-supervision that matches a user question with a trusted medical knowledge base and retrieves a fixed number of relevant sentences from the corresponding answer document. |
| Outcome: | The proposed system retrieves more relevant answers while achieving 20 times faster. |
Characterizing Departures from Linearity in Word Translation (P18-2)
Copied to clipboard
| Challenge: | a class of methods has emerged to learn linear maps between word embedding spaces of different languages. |
| Approach: | They propose to approximate word embedding spaces using linear maps . they show that the underlying maps are non-linear but vary by a proportion of distance . |
| Outcome: | The proposed methods can be used to test non-linear methods and drive the design of more accurate maps for word translation. |
SYMPTOMIFY: Transforming Symptom Annotations with Language Model Knowledge Harvesting (2023.findings-emnlp)
Copied to clipboard
| Challenge: | a new dataset of annotated vaccine adverse reaction reports is aimed at improving human annotators . a continual evolution in language models and strides in few-shot learning offer promise for improvement. |
| Approach: | They propose a resource to help human annotators improve their efficiency . they evaluate performance across various methods and learning paradigms . |
| Outcome: | The proposed resource outperforms existing systems and learning paradigms in evaluating their performance. |
Commonsense about Human Senses: Labeled Data Collection Processes (D19-60)
Copied to clipboard
| Challenge: | Existing methods for recognizing mentions of human senses in text are lacking in common sense knowledge acquisition. |
| Approach: | They propose to use machine learning to acquire labeled data to extract common sense relationships pertaining to sense perception concepts. |
| Outcome: | The proposed method is effective when used with standard machine learning models on the task of sense recognition in text. |
NORMA: Neighborhood Sensitive Maps for Multilingual Word Embeddings (D18-1)
Copied to clipboard
| Challenge: | Existing methods for learning multilingual word embeddings assume that embeddable spaces of different languages exhibit similar structures. |
| Approach: | They propose a method for learning neighborhood sensitive maps to capture such differences . aim is to learn word vectors where similar words have similar vector representations . |
| Outcome: | The proposed method outperforms state-of-the-art methods for translation between distant languages. |
Recursive Tree-Structured Self-Attention for Answer Sentence Selection (2021.acl-long)
Copied to clipboard
| Challenge: | Recent top-performing models in Answer Sentence Selection use self-attention and transfer learning, but not syntactic structure. |
| Approach: | They propose a recursive, tree-structured self-attention model that can represent all levels of syntactic parse trees with only one additional layer. |
| Outcome: | The proposed model can represent all levels of syntactic parse trees with only one additional layer without transfer learning. |
Fine-Grained Spoiler Detection from Large-Scale Review Corpora (P19-1)
Copied to clipboard
| Challenge: | 'Spoilers' on review websites can be a concern for consumers who want to fully experience the excitement of media consumption. |
| Approach: | They propose to use a large-scale book review dataset to generate fine-grained spoiler annotations . they then use supervised neural networks to detect spoiler sentences in review corpora . |
| Outcome: | The proposed method outperforms baselines in a large-scale book review dataset . it can detect spoiler sentences in review corpora, but only a few users use it . |
On Linearizing Structured Data in Encoder-Decoder Language Models: Insights from Text-to-SQL (2024.naacl-long)
Copied to clipboard
| Challenge: | Structured data is prevalent in tables, databases, and knowledge graphs, but there is a gap in our understanding of how these linearization-based methods handle structured data, which is inherently non-linear. |
| Approach: | They investigate the linear handling of structured data in encoder-decoder language models, specifically T5. |
| Outcome: | The proposed model can mimic human-designed processes such as schema linking and syntax prediction, and it can be compressed due to modality fusion redundancy. |
ChartDialogs: Plotting from Natural Language Instructions (2020.acl-main)
Copied to clipboard
| Challenge: | a new dataset of conversational plotting agents is developed to facilitate the development of such agents. |
| Approach: | They propose a dataset that contains over 15,000 dialog turns from matplotlib's most popular plotting library. |
| Outcome: | The proposed system achieves 61% plotting accuracy, compared to the previous method. |
Interactive Plot Manipulation using Natural Language (2021.naacl-demos)
Copied to clipboard
| Challenge: | a new interactive plotting agent is available for programming with natural language . the interactive aspect allows users to manipulate plots using natural language instructions. |
| Approach: | They propose an interactive natural language interface for plotting that maps language to plot updates. |
| Outcome: | The proposed system maps language to plot updates within an interactive programming environment. |
A Gradually Soft Multi-Task and Data-Augmented Approach to Medical Question Understanding (2021.acl-long)
Copied to clipboard
Khalil Mrini, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter Chang, Emilia Farcas, Ndapa Nakashole
| Challenge: | Existing methods for medical question understanding often fail to provide high recall in answer retrieval. |
| Approach: | They propose a multi-task learning method with data augmentation for medical question understanding that uses just one dataset to optimize for both tasks. |
| Outcome: | The proposed method outperforms existing MTL methods across 4 datasets of medical question pairs in ROUGE scores, RQE accuracy and human evaluation. |
Grammar as Control: Modular Language Generation for the Long Tail (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) can bootstrap language technologies for long-tail languages . however, without structured guidance, they produce narrow, unrepresentative samples . |
| Approach: | They propose a framework that transforms descriptive grammars into explicit control mechanisms that guide LLMs to generate typologically balanced synthetic data for downstream training. |
| Outcome: | The proposed framework improves typological entropy and yields a "student-beats-teacher" effect across three low-resource languages. |