DAFMSVC: One-Shot Singing Voice Conversion with Dual Attention Mechanism and Flow Matching

Kavli Affiliate: Dan Luo | First 5 Authors: Wei Chen, Wei Chen, , , | Summary: Singing Voice Conversion (SVC) transfers a source singer’s timbre to a target while keeping melody and lyrics. The key challenge in any-to-any SVC is adapting unseen speaker timbres to source audio without quality degradation. Existing methods either face timbre […]


Continue.. DAFMSVC: One-Shot Singing Voice Conversion with Dual Attention Mechanism and Flow Matching

Computational Design and Fabrication of Modular Robots with Untethered Control

Kavli Affiliate: Chiara Daraio | First 5 Authors: Manas Bhargava, Manas Bhargava, , , | Summary: Natural organisms use distributed actuation via their musculoskeletal systems to adapt their gait for traversing diverse terrains or to morph their bodies to perform varied tasks. A longstanding challenge in the field of robotics is to mimic this extensive […]


Continue.. Computational Design and Fabrication of Modular Robots with Untethered Control

Computational Design and Fabrication of Modular Robots with Untethered Control

Kavli Affiliate: Chiara Daraio | First 5 Authors: Manas Bhargava, Manas Bhargava, , , | Summary: Natural organisms utilize distributed actuation through their musculoskeletal systems to adapt their gait for traversing diverse terrains or to morph their bodies for varied tasks. A longstanding challenge in robotics is to emulate this capability of natural organisms, which […]


Continue.. Computational Design and Fabrication of Modular Robots with Untethered Control

Experimental demonstration of corrugated nanolaminate films as reflective light sails

Kavli Affiliate: Harry A. Atwater | First 5 Authors: Matthew F. Campbell, Matthew F. Campbell, , , | Summary: Achieving laser-driven, reflective, relativistic light sails would represent a tremendous breakthrough for humankind, allowing us to advance our understanding of the solar system and deep space far beyond what we know from space probes, telescopes, and […]


Continue.. Experimental demonstration of corrugated nanolaminate films as reflective light sails

Model Accuracy and Data Heterogeneity Shape Uncertainty Quantification in Machine Learning Interatomic Potentials

Kavli Affiliate: Wei Gao | First 5 Authors: Fei Shuang, Fei Shuang, , , | Summary: Machine learning interatomic potentials (MLIPs) enable accurate atomistic modelling, but reliable uncertainty quantification (UQ) remains elusive. In this study, we investigate two UQ strategies, ensemble learning and D-optimality, within the atomic cluster expansion framework. It is revealed that higher […]


Continue.. Model Accuracy and Data Heterogeneity Shape Uncertainty Quantification in Machine Learning Interatomic Potentials

Simulating high-temperature superconductivity in moiré WSe2

Kavli Affiliate: Jie Shan | First 5 Authors: Yiyu Xia, Yiyu Xia, , , | Summary: The emergence of high transition temperature (Tc) superconductivity in strongly correlated materials remains a major unsolved problem in physics. High-Tc materials, such as cuprates, are generally complex and not easily tunable, making theoretical modelling difficult. Although the Hubbard model–a […]


Continue.. Simulating high-temperature superconductivity in moiré WSe2

FastCSP: Accelerated Molecular Crystal Structure Prediction with Universal Model for Atoms

Kavli Affiliate: Chiara Daraio | First 5 Authors: Vahe Gharakhanyan, Vahe Gharakhanyan, , , | Summary: Crystal Structure Prediction (CSP) of molecular crystals plays a central role in applications, such as pharmaceuticals and organic electronics. CSP is challenging and computationally expensive due to the need to explore a large search space with sufficient accuracy to […]


Continue.. FastCSP: Accelerated Molecular Crystal Structure Prediction with Universal Model for Atoms

MMBERT: Scaled Mixture-of-Experts Multimodal BERT for Robust Chinese Hate Speech Detection under Cloaking Perturbations

Kavli Affiliate: Wei Gao | First 5 Authors: Qiyao Xue, Qiyao Xue, , , | Summary: Hate speech detection on Chinese social networks presents distinct challenges, particularly due to the widespread use of cloaking techniques designed to evade conventional text-based detection systems. Although large language models (LLMs) have recently improved hate speech detection capabilities, the […]


Continue.. MMBERT: Scaled Mixture-of-Experts Multimodal BERT for Robust Chinese Hate Speech Detection under Cloaking Perturbations

YOLO-Count: Differentiable Object Counting for Text-to-Image Generation

Kavli Affiliate: Xiang Zhang | First 5 Authors: Guanning Zeng, Guanning Zeng, , , | Summary: We propose YOLO-Count, a differentiable open-vocabulary object counting model that tackles both general counting challenges and enables precise quantity control for text-to-image (T2I) generation. A core contribution is the ‘cardinality’ map, a novel regression target that accounts for variations […]


Continue.. YOLO-Count: Differentiable Object Counting for Text-to-Image Generation

Information-Theoretic Decentralized Secure Aggregation with Collusion Resilience

Kavli Affiliate: Xiang Zhang | First 5 Authors: Xiang Zhang, Xiang Zhang, , , | Summary: In decentralized federated learning (FL), multiple clients collaboratively learn a shared machine learning (ML) model by leveraging their privately held datasets distributed across the network, through interactive exchange of the intermediate model updates. To ensure data security, cryptographic techniques […]


Continue.. Information-Theoretic Decentralized Secure Aggregation with Collusion Resilience