Three papers by TRAILS authors accepted to NAACL 2025 and associated workshops, and one to COLING 2025

Two papers originating from research in the TRAILS project have been accepted to the Main Track and Findings of the 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL). NAACL will take place April 29–May 4 in Albuquerque, New Mexico.

The first paper proposes soft language prompts for improved cross-lingual transfer, outperforming traditional methods, especially in low-resource settings. In the second paper, we present a repository of static word embeddings for 87 low-resource languages, enhanced by integrating multilingual graph knowledge to improve performance in natural language processing tasks, performing at par with LLMs in some settings.

An additional paper has been accepted to the 10th Workshop on Representation Learning for NLP (RepL4NLP), co-located with NAACL 2025. The paper reimagines classical probing to evaluate knowledge transfer from simple source to more complex target tasks. Instead of probing frozen representations from a complex source task on diverse simple target probing tasks (as usually done in probing), we explore the effectiveness of embeddings from multiple simple source tasks on a single target task.

One paper was accepted at the 31st International Conference on Computational Linguistics 2025 (COLING 2025). In the paper, we present CROSS-REFINE, a generator-critic framework that enhances natural language explanations by refining initial outputs using feedback from a second LLM, outperforming SELF-REFINE and working effectively even with less powerful models.