Papers by Nicolas Le Roux
fLSA: Learning Semantic Structures in Document Collections Using Foundation Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) can be used to solve new tasks by inducing high-level strategies from example solutions to similar problems and adapting these strategies to solve unseen problems. |
| Approach: | They propose a foundation-model-based Latent Semantic Analysis method that iteratively clusters and tags document segments based on document-level contexts. |
| Outcome: | The proposed method is more informative in reconstructing the original texts than existing tagging methods. |