S2ST-Omni: An Efficient Multilingual Speech-to-Speech Translation Framework via Seamless Speech-Text Alignment and Progressive Fine-tuning

Posted by klaurent Berkeley

Kavli Affiliate: Xiang Zhang | First 5 Authors: Yu Pan, Yu Pan, , , | Summary: Despite recent advances in multilingual speech-to-speech translation (S2ST), several critical challenges persist: 1) achieving high-quality translation remains a major hurdle, and 2) most existing methods heavily rely on large-scale parallel speech corpora, which are costly and difficult to obtain. […]


Continue.. S2ST-Omni: An Efficient Multilingual Speech-to-Speech Translation Framework via Seamless Speech-Text Alignment and Progressive Fine-tuning

Optical spin pumping in silicon

Posted by dbos Delft

Kavli Affiliate: Giordano Scappucci | First 5 Authors: Stefano Achilli, Damiano Marian, Mario Lodari, Emiliano Bonera, Giordano Scappucci | Summary: The generation of an out-of-equilibrium population of spin-polarized carriers is a keystone process for quantum technologies and spintronics alike. It can be achieved through the so-called optical spin orientation by exciting the material with circularly […]


Continue.. Optical spin pumping in silicon

Generalized Gaussian Entropy Model for Point Cloud Attribute Compression with Dynamic Likelihood Intervals

Posted by dbos Caltech

Kavli Affiliate: Wei Gao | First 5 Authors: Changhao Peng, Yuqi Ye, Wei Gao, , | Summary: Gaussian and Laplacian entropy models are proved effective in learned point cloud attribute compression, as they assist in arithmetic coding of latents. However, we demonstrate through experiments that there is still unutilized information in entropy parameters estimated by […]


Continue.. Generalized Gaussian Entropy Model for Point Cloud Attribute Compression with Dynamic Likelihood Intervals

Enhancing quantum noise characterization via extra energy levels

Posted by dbos Berkeley

Kavli Affiliate: Irfan Siddiqi | First 5 Authors: Senrui Chen, Akel Hashim, Noah Goss, Alireza Seif, Irfan Siddiqi | Summary: Noise is a major challenge for building practical quantum computing systems. Precise characterization of quantum noise is crucial for developing effective error mitigation and correction schemes. However, state preparation and measurement (SPAM) errors on many […]


Continue.. Enhancing quantum noise characterization via extra energy levels

Enhancing quantum noise characterization via extra energy levels

Posted by klaurent Berkeley

Kavli Affiliate: Irfan Siddiqi | First 5 Authors: Senrui Chen, Senrui Chen, , , | Summary: Noise is a major challenge for building practical quantum computing systems. Precise characterization of quantum noise is crucial for developing effective error mitigation and correction schemes. However, state preparation and measurement (SPAM) errors on many current platforms can introduce […]


Continue.. Enhancing quantum noise characterization via extra energy levels

MagCache: Fast Video Generation with Magnitude-Aware Cache

Posted by dbos Berkeley

Kavli Affiliate: Feng Wang | First 5 Authors: Zehong Ma, Longhui Wei, Feng Wang, Shiliang Zhang, Qi Tian | Summary: Existing acceleration techniques for video diffusion models often rely on uniform heuristics or time-embedding variants to skip timesteps and reuse cached features. These approaches typically require extensive calibration with curated prompts and risk inconsistent outputs […]


Continue.. MagCache: Fast Video Generation with Magnitude-Aware Cache

Site-resolved magnon and triplon dynamics on a programmable quantum dot spin ladder

Posted by klaurent Delft

Kavli Affiliate: Menno Veldhorst | First 5 Authors: Pablo Cova Fariña, Pablo Cova Fariña, , , | Summary: Quasi-particle dynamics in interacting systems in the presence of disorder challenges the notion of internal thermalization, but proves difficult to investigate theoretically for large particle numbers. Engineered quantum systems may offer a viable alternative, as witnessed in […]


Continue.. Site-resolved magnon and triplon dynamics on a programmable quantum dot spin ladder

Broadband and high-precision two-level system loss measurement using superconducting multi-wave resonators

Posted by dbos Caltech

Kavli Affiliate: Oskar Painter | First 5 Authors: Cliff Chen, Shahriar Aghaeimeibodi, Yuki Sato, Matthew H. Matheny, Oskar Painter | Summary: Two-level systems (TLS) are known to be a dominant source of dissipation and decoherence in superconducting qubits. Superconducting resonators provide a convenient way to study TLS-induced loss due to easier design and fabrication in […]


Continue.. Broadband and high-precision two-level system loss measurement using superconducting multi-wave resonators

Reinforcement Learning Optimization for Large-Scale Learning: An Efficient and User-Friendly Scaling Library

Posted by dbos Caltech

Kavli Affiliate: Wei Gao | First 5 Authors: Weixun Wang, Shaopan Xiong, Gengru Chen, Wei Gao, Sheng Guo | Summary: We introduce ROLL, an efficient, scalable, and user-friendly library designed for Reinforcement Learning Optimization for Large-scale Learning. ROLL caters to three primary user groups: tech pioneers aiming for cost-effective, fault-tolerant large-scale training, developers requiring flexible […]


Continue.. Reinforcement Learning Optimization for Large-Scale Learning: An Efficient and User-Friendly Scaling Library