Papers with word-level representations
WACO: Word-Aligned Contrastive Learning for Speech Translation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing ST methods perform poorly when only a limited amount of parallel data are available for training. |
| Approach: | They propose a Word-Aligned COntrastive learning method for low-resource speech-to-text translation that bridges word-level representations for both speech and text modalities via contrastive learning. |
| Outcome: | The proposed method outperforms the best baseline by 9+ BLEU points with only 1-hour parallel ST data. |
Towards end-2-end learning for predicting behavior codes from spoken utterances in psychotherapy conversations (2020.acl-main)
Copied to clipboard
| Challenge: | Xu and Sarikaya, 2014) proposes a framework for predicting utterance level labels directly from speech features. |
| Approach: | They propose a framework for predicting utterance level labels directly from speech features using a pretrained Speech-2-Vector encoder as bottleneck. |
| Outcome: | The proposed model outperforms state-of-the-art approaches which use transcribed text for the task of predicting psychotherapy-relevant behavior codes. |
Learning Decoupled Retrieval Representation for Nearest Neighbour Neural Machine Translation (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods to integrate external corpus are sparse in practical applications, and noises in low similarity retrieval could lead to severe performance degradation. |
| Approach: | They propose a method to integrate external corpus into k-nearest neighbor machine translation (kNNMT) instead of storing discrete word sequence, kNN-MT uses a pre-trained NMT model to force decoding the external corpi. |
| Outcome: | The proposed approach improves retrieval accuracy and BLEU score on five domains compared to vanilla kNNMT. |
Understanding Subword Compositionality of Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) take sequences of subwords as input, requiring them to compose subword representations into meaningful word-level representations. |
| Approach: | They propose to probe how large language models compose subword information . they find structural similarity, semantic decomposability, and form retention are key aspects . |
| Outcome: | The proposed models can be classified into three distinct groups, the authors show . they show that they can achieve great performance when probing layer by layer their sensitivity to semantic decompositionality . |
Word-level Commonsense Knowledge Selection for Event Detection (2024.lrec-main)
Copied to clipboard
| Challenge: | Event Detection (ED) is a task of automatically extracting multi-class trigger words . Xie and Tu, 2022, use a Context-specific Knowledge Selector to select commonsense knowledge of words based on living contexts . |
| Approach: | They use a Context-specific Knowledge Selector to select the exact commonsense knowledge of words from a large knowledge base. |
| Outcome: | The proposed approach achieves the F1-score of about 78.3% on the ACE-2005 dataset. |