Papers by Julian Risch
Multifaceted Domain-Specific Document Embeddings (2021.naacl-demos)
Copied to clipboard
| Challenge: | Current document embeddings require large training corpora but fail to learn high-quality representations when confronted with a small number of domain-specific documents and rare terms. |
| Approach: | They propose a faceted domain encoder that transforms each document into a single embedding vector . they use a Siamese neural network architecture to leverage knowledge graphs to enhance the embeddables . |
| Outcome: | The proposed model achieves the same embedding quality as state-of-the-art models while requiring only a tiny fraction of training data. |
Prediction for the Newsroom: Which Articles Will Get the Most Comments? (N18-3)
Copied to clipboard
| Challenge: | a new method to support manual moderation of discussion sections is proposed. |
| Approach: | They propose to support manual moderation by proactively drawing attention of moderators to articles that most likely need their intervention. |
| Outcome: | The proposed method outperforms the current state-of-the-art methods on a 7-million-comment dataset. |
Fabricator: An Open Source Toolkit for Generating Labeled Training Data with Teacher LLMs (2023.emnlp-demo)
Copied to clipboard
| Challenge: | Recent research addresses the bottleneck of producing labeled training data for NLP tasks. |
| Approach: | They propose a method that generates labeled data that can be used to train a downstream NLP model. |
| Outcome: | The proposed model enables an LLM to generate labeled data that can be used to train a downstream NLP model. |