Reusable skills are becoming a common interface for extending large language model agents, packaging procedural guidance with access to files, tools, memory, and execution environments. However, this modularity introduces attack surfaces that are largely missed by existing safety evaluations: even when the user request is benign, unsafe influence may reside in skill guidance, local artifacts, or execution-environment files that steer the agent toward unsafe actions. We present **SkillSafetyBench**, a runnable benchmark for evaluating such skill-mediated safety failures. SkillSafetyBench includes 155 adversarial cases across 47 tasks, 6 risk domains, and 30 safety categories, each evaluated with a case-specific rule-based verifier. Experiments with multiple CLI agents and model backends show that non-user attacks can consistently induce unsafe behavior, with distinct failure patterns across domains, attack methods, and scaffold–model pairings. Our findings suggest that agent safety depends not only on model-level alignment, but also on how agents interpret skills, trust workflow context, and act through executable environments. We release the data and code in an anonymous repository to support reproducibility. The complete benchmark is available at https://github.com/AI45Lab/skill-safety-bench.

Manifold’s English-Chinese System at WMT22 is an ensemble of 4 models trained by different configurations with scheduled sampling-based fine-tuning. The four configurations are DeepBig (XenC), DeepLarger (XenC), DeepBig-TalkingHeads (XenC) and DeepBig (LaBSE). Concretely, DeepBig extends Transformer-Big to 24 encoder layers. DeepLarger has 20 encoder layers and its feed-forward network (FFN) dimension is 8192. TalkingHeads applies the talking-heads trick. For XenC configs, we selected monolingual and parallel data that is similar to the past newstest datasets using XenC, and for LaBSE, we cleaned the officially provided parallel data using LaBSE pretrained model. According to the officially released autonomic metrics leaderboard, our final constrained system ranked 1st among all others when evaluated by bleu-all, chrf-all and COMET-B, 2nd by COMET-A.

神经机器翻译(NMT)通常采用多层神经网络模型结构,随着网络层数的加深,所得到的特征也越来越抽象,但是在现有的神经机器翻译模型中,高层的抽象信息仅在预测分布时被利用。为了更好地利用这些信息,本文提出了层间知识蒸馏,目的在于将高层网络的抽象知识迁移到低层网络,使低层网络能够捕捉更加有用的信息,从而提升整个模型的翻译质量。区别于传统教师模型和学生模型的知识蒸馏,层间知识蒸馏实现的是同一个模型内部不同层之间的知识迁移。通过在中文-英语、英语-罗马尼亚语、德语-英语三个数据集上的实验,结果证明层间蒸馏方法能够有效提升翻译性能,分别在中-英、英-罗、德-英上提升1.19,0.72,1.35的BLEU值,同时也证明有效地利用高层信息能够提高神经网络模型的翻译质量。