The effects of tuna viscera hydrolysate (TVH) on juvenile pompano Trachinotus blochii, growth performance, nutritional response, intestinal and liver health, and resistance to Streptococcus iniae were investigated in this study. Five isonitrogenous and isocaloric diets (protein 46.0%, lipid 10.0%) were formulated in which TVH was added to replace fishmeal protein at levels of 0 (control), 30, 60, 90, and 120 g kg-1, labelled as TVH0, TVH05, TVH10, TVH15, and TVH20, respectively. Triplicate groups of pompano were fed the respective diets for ten weeks. The results showed that fish fed diets containing TVH10 produced significantly higher final body weight and specific growth rate in comparison to the fishmeal control ( P < 0.05 ). Dietary TVH did not produce any effect on feed utilisation, somatic indices, and proximate composition of juvenile pompano ( P > 0.05 ). While most amino acids were unchanged by the dietary inclusion of TVH, phenylalanine and valine levels were significantly lower in the fish fed TVH20 diet compared to the control. Fish fed the TVH20 diet had significantly lowered total serum protein compared to the TVH10 treatment, whereas other biochemical parameters in the blood did not show any difference among treatments. The intestinal histology indicated a significant increase in goblet cell numbers in fish fed TVH10 diet. Fish fed diet supplemented with TVH showed the highest disease resistance against Streptococcus iniae after 14 days of challenge. Based on a quadratic regression between final body weight and dietary TVH levels, the optimum TVH was calculated to be 10% or 60.0 g kg-1 for maximum growth performance when fed to pompano.

In this work, we focus on Cross-Lingual Event Detection where a model is trained on data from a source language but its performance is evaluated on data from a second, target, language. Most recent works in this area have harnessed the language-invariant qualities displayed by pre-trained Multi-lingual Language Models. Their performance, however, reveals there is room for improvement as the crosslingual setting entails particular challenges. We employ Adversarial Language Adaptation to train a Language Discriminator to discern between the source and target languages using unlabeled data. The discriminator is trained in an adversarial manner so that the encoder learns to produce refined, language-invariant representations that lead to improved performance. More importantly, we optimize the adversarial training process by only presenting the discriminator with the most informative samples. We base our intuition about what makes a sample informative on two disparate metrics: sample similarity and event presence. Thus, we propose leveraging Optimal Transport as a solution to naturally combine these two distinct information sources into the selection process. Extensive experiments on 8 different language pairs, using 4 languages from unrelated families, show the flexibility and effectiveness of our model that achieves state-of-the-art results.

No abstract

The cover image, by Minh Van Nguyen et al., is based on the original article The role of dietary methionine concentrations on growth, metabolism and N‐retention in cobia (Rachycentron canadum) at elevated water temperatures. DOI: .

No abstract