The MEDIQA-EVAL 2026 shared task focuses on developing automatic evaluation metrics for LLM-generated responses in dermatology and wound care. While LLMs have shown promise as judge models, the reliability of these metrics remains underexplored. In this work, we study how well judge models can approximate human expert ratings across clinical evaluation criteria. We evaluate multiple approaches, including few-shot prompting, BERT fine-tuning, and retrieval-augmented generation (RAG), and combine them in an ensemble framework. Our method achieves a correlation score of 0.481, ranking first among 41 participating teams. Our results provide insight into the reliability of LLM-based evaluation metrics and highlight their potential for scalable clinical assessment.

We consider the task of cross-lingual adaptation of dependency parsers without annotated target corpora and parallel corpora. Previous work either directly applies a discriminative source parser to the target language, ignoring unannotated target corpora, or employs an unsupervised generative parser that can leverage unannotated target data but has weaker representational power than discriminative parsers. In this paper, we propose to utilize unsupervised discriminative parsers based on the CRF autoencoder framework for this task. We train a source parser and use it to initialize and regularize a target parser that is trained on unannotated target data. We conduct experiments that transfer an English parser to 20 target languages. The results show that our method significantly outperforms previous methods.

Nowadays, more and more customers browse and purchase products in favor of using mobile E-Commerce Apps such as Taobao and Amazon. Since merchants are usually inclined to describe redundant and over-informative product titles to attract attentions from customers, it is important to concisely display short product titles on limited screen of mobile phones. To address this discrepancy, previous studies mainly consider textual information of long product titles and lacks of human-like view during training and evaluation process. In this paper, we propose a Multi-Modal Generative Adversarial Network (MM-GAN) for short product title generation in E-Commerce, which innovatively incorporates image information and attribute tags from product, as well as textual information from original long titles. MM-GAN poses short title generation as a reinforcement learning process, where the generated titles are evaluated by the discriminator in a human-like view. Extensive experiments on a large-scale E-Commerce dataset demonstrate that our algorithm outperforms other state-of-the-art methods. Moreover, we deploy our model into a real-world online E-Commerce environment and effectively boost the performance of click through rate and click conversion rate by 1.66% and 1.87%, respectively.