Adaptive FSS: A Novel Few-Shot Segmentation Framework via Prototype Enhancement

Kavli Affiliate: Jing Wang | First 5 Authors: Jing Wang, Jinagyun Li, Chen Chen, Yisi Zhang, Haoran Shen | Summary: The Few-Shot Segmentation (FSS) aims to accomplish the novel class segmentation task with a few annotated images. Current FSS research based on meta-learning focus on designing a complex interaction mechanism between the query and support […]


Continue.. Adaptive FSS: A Novel Few-Shot Segmentation Framework via Prototype Enhancement

Revisiting Metastable Cosmic String Breaking

Kavli Affiliate: Satoshi Shirai | First 5 Authors: Akifumi Chitose, Masahiro Ibe, Yuhei Nakayama, Satoshi Shirai, Keiichi Watanabe | Summary: Metastable cosmic strings appear in models of new physics with a two-step symmetry breaking $Gto Hto 1$, where $pi_1(H)neq 0$ and $pi_1(G)=0$. They decay via the monopole-antimonopole pair creation inside. Conventionally, the breaking rate has […]


Continue.. Revisiting Metastable Cosmic String Breaking

Revisiting Metastable Cosmic String Breaking

Kavli Affiliate: Satoshi Shirai | First 5 Authors: Akifumi Chitose, Masahiro Ibe, Yuhei Nakayama, Satoshi Shirai, Keiichi Watanabe | Summary: Metastable cosmic strings appear in models of new physics with a two-step symmetry breaking $Gto Hto 1$, where $pi_1(H)neq 0$ and $pi_1(G)=0$. They decay via the monopole-antimonopole pair creation inside. Conventionally, the breaking rate has […]


Continue.. Revisiting Metastable Cosmic String Breaking

Balanced SNR-Aware Distillation for Guided Text-to-Audio Generation

Kavli Affiliate: Yi Zhou | First 5 Authors: Bingzhi Liu, Yin Cao, Haohe Liu, Yi Zhou, | Summary: Diffusion models have demonstrated promising results in text-to-audio generation tasks. However, their practical usability is hindered by slow sampling speeds, limiting their applicability in high-throughput scenarios. To address this challenge, progressive distillation methods have been effective in […]


Continue.. Balanced SNR-Aware Distillation for Guided Text-to-Audio Generation