Papers by Junya Takayama
Text Classification with Negative Supervision (2020.acl-main)
Copied to clipboard
| Challenge: | Existing models for text representations have shown state-of-the-art performance on text classification tasks, however, the discrepancy between semantic similarity of texts and labelling standards affects classifiers. |
| Approach: | They propose a simple multitask learning model that uses negative supervision to generate distinct representations for texts with different labels. |
| Outcome: | The proposed model outperforms state-of-the-art models on classification tasks in three different languages. |
Persona-Consistent Dialogue Generation via Pseudo Preference Tuning (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for improving persona consistency in dialogues require external resources. |
| Approach: | They propose a method for enhancing persona consistency in dialogue response generation using direct preference optimization using persona data. |
| Outcome: | The proposed method produces more consistent and natural responses than previous methods. |
DIRECT: Direct and Indirect Responses in Conversational Text Corpus (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Neural conversation models have been able to generate fluent responses through training on a dialogue corpus, but they lack the ability to reveal the implied intentions of users. |
| Approach: | They propose to train neural conversation models on a dialogue corpus that provides pragmatic paraphrases to advance techniques for natural language understanding in dialogue systems. |
| Outcome: | The proposed corpus provides 71,498 pairs of indirect–direct utterance pairs accompanied by a multi-turn dialogue history extracted from the MultiWoZ dataset. |
Dialogue-Act Prediction of Future Responses Based on Conversation History (P19-2)
Copied to clipboard
| Challenge: | Sequence-to-sequence models are a common approach to develop chatbots, but they are prone to a black-box response generation process. |
| Approach: | They propose a method to predict a DA of the next response based on the history of previous utterances and their DAs. |
| Outcome: | The proposed model achieves 10.8% higher F1-score and 3.0% higher accuracy on DA prediction compared to baseline using only a single utterance . |
Distinct Label Representations for Few-Shot Text Classification (2021.acl-short)
Copied to clipboard
| Challenge: | Existing methods for few-shot text classification ignore the semantic relevance of labels and are difficult to train because of the lack of training examples. |
| Approach: | They propose a method that generates distinct label representations that embed information specific to each label. |
| Outcome: | The proposed method significantly improves few-shot text classification across models and datasets. |
Consistent Response Generation with Controlled Specificity (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to generate fluent responses generate inconsistent responses . we use a sequence-to-sequence model to generate specific responses based on a co-occurrence degree . |
| Approach: | They propose a method to control the specificity of responses while maintaining the consistency with the utterances. |
| Outcome: | The proposed method produces highly consistent responses in open-domain dialogues . it can generate fluent responses while maintaining the consistency with the utterances compared to the conventional model . |