K-Bench is introduced, a benchmark that scores LLM unlearning under agentic deployment and certifies forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten.
Guang-Sheng Yu, Yan-Na Jiang, Qin Wang et al.· 0 citations
Unlearning benchmarks such as TOFU and MUSE certify forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten. We show that this model-level certificate does not transfer once the model is deployed as an agent. We introduce K-Bench, a benchmark that scores L...
Guangsheng Yu, Yanna Jiang, Qin Wang et al.· 0 citations
This review analyzes the potential contribution of adjuvant strategies to developing a universal porcine rotavirus vaccine capable of broad, long-lasting cross-protection and identifies key unresolved scientific questions and critical knowledge gaps in the field with the aim of guiding future research priorities.
Zhen Zhang, Baihe Ma, Hong-Miao Zhou et al.· Frontiers in Veterinary Scie...· 0 citations
The results support a model in which S. parasuis accumulates AMR-related genes in a modular fashion via mobile elements within stable species boundaries, with no detectable same-MGE colocalization of AMR and virulence determinants under the analytical pipeline.
A novel NSP2 deletion pattern and critical amino acid substitutions in the GP5 protein were identified and are consistent with previous regional molecular epidemiological surveillance and may serve as a basis for broader investigations into the genetic evolution of PRRSV.
Shuhua Liu, Zhen Zhang, Baihe Ma et al.· Microbial Pathogenesis· 0 citations
This review systematically synthesizes key advances over the past decade concerning host-Eimeria interactions, molecular regulatory mechanisms, and novel control strategies, while contextualizing these findings with earlier seminal discoveries.