ALT Bregman Conditional Random Fields: Sequence Labeling with Parallelizable Inference Algorithms (Caio Corro, Mathieu Lacroix, Joseph Le Roux)

ALT Few-Shot Domain Adaptation for Named-Entity Recognition via Joint Constrained k-Means and Subspace Selection Ayoub Hammal, Benno Uthayasooriyar, Caio Corro

ALT Named-entity recognition (NER) is a task that typically requires large annotated datasets, which limits its applicability across domains with varying entity definitions. This paper addresses few-shot NER, aiming to transfer knowledge to new domains with minimal su- pervision. Unlike previous approaches that rely solely on limited annotated data, we propose a weakly supervised algorithm that combines small labeled datasets with large amounts of unlabeled data. Our method extends the k-means algorithm with label supervision, cluster size constraints and domain-specific discriminative subspace selection. This unified frame work achieves state-of-the-art results in few-shot NER on several English datasets.

ALT Train process illustration.

ALT You are requested NOT to use the response form to debate the reviewers' subjective opinions regarding the merit of your work. Nor should you try to "correct" your paper in any way - either in terms of its basic technical arguments, or in the presentation of those arguments. Above all, the response facility should not be used to report on new results, obtained since the submission deadline closed. Of course, you are free to write whatever you wish to write. That said, please note that this process is intended to help produce more accurate REVIEWS for your paper. It is not intended to strengthen the arguments in your paper; moreover, your remarks will not lead to changing the reviewers' fundamental perceptions of your work, as reflected in their reviews.

ALT Sparse Logistic Regression with High-order Features for Automatic Grammar Rule Extraction from Treebanks Santiago Herrera, Caio Corro, Sylvain Kahane Descriptive grammars are highly valuable, but writing them is time-consuming and difficult. Furthermore, while linguists typically use corpora to create them, grammar descriptions often lack quantitative data. As for formal grammars, they can be challenging to interpret. In this paper, we propose a new method to extract and explore significant fine-grained grammar patterns and potential syntactic grammar rules from treebanks, in order to create an easy-to-understand corpus-based grammar. More specifically, we extract descriptions and rules across different languages for two linguistic phenomena, agreement and word order, using a large search space and paying special attention to the ranking order of the extracted rules. For that, we use a linear classifier to extract the most salient features that predict the linguistic phenomena u