Papers by Philip Cohen
Improving Cross-Domain Low-Resource Text Generation through LLM Post-Editing: A Programmer-Interpreter Approach (2024.findings-eacl)
Copied to clipboard
| Challenge: | Large pre-trained language models such as GPT-3.5 and GPT-4 have gained significant attention in natural language research due to limited computational resources or inaccessible parameters. |
| Approach: | They propose a neural programmer-interpreter approach that preserves the domain generalization ability of LLMs while editing their output. |
| Outcome: | The proposed framework significantly improves GPT-3.5’s performance in logical form-to-text conversion and low-resource machine translation, surpassing other state-of-the-art (SOTA) LLM post-editing methods in cross-domain settings. |
The Best of Both Worlds: Combining Human and Machine Translations for Multilingual Semantic Parsing with Active Learning (2023.acl-long)
Copied to clipboard
| Challenge: | Prior studies have focused on translating utterances from high-resource languages to low-resourced languages. |
| Approach: | They propose an active learning approach that exploits the strengths of both human and machine translations by iteratively adding small batches of human translations into the machine-translated training set. |
| Outcome: | The proposed approach reduces errors and bias in the translated data, resulting in higher parser accuracies than the current model trained on machine translations. |
Active learning for deep semantic parsing (P18-2)
Copied to clipboard
| Challenge: | Existing methods for generating training data for semantic parsing are slow and expensive. |
| Approach: | They propose active learning for "overnight" and "natural language" parsing with a logical form . they propose several active learning strategies for overnight data collection . |
| Outcome: | The proposed approach reduces the cost of training data for deep parsing tasks by reducing the number of crowd workers required. |