KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device Inference

Kavli Affiliate: Huawei Zhang | Summary:Language models (LMs) underpin emerging mobile and embedded AI applications like meeting and video summarization and document analysis, which often require processing multiple long-context inputs. Running an LM locally on-device improves privacy, enables offline use, and reduces cost, but long-context inference quickly hits a emphmemory capacity wall as the key-value […]


Continue.. KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device Inference

KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device Inference

Kavli Affiliate: Huawei Zhang| First 5 Authors: Huawei Zhang, Huawei Zhang, , , | Summary:Language models (LMs) underpin emerging mobile and embedded AI applications like meeting and video summarization and document analysis, which often require processing multiple long-context inputs. Running an LM locally on-device improves privacy, enables offline use, and reduces cost, but long-context inference […]


Continue.. KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device Inference

Measurement of reactor antineutrino oscillations with 1.46 ktonne-years of data at SNO+

Kavli Affiliate: N. C. Yeh| First 5 Authors: [#item_custom_name[1, [#item_custom_name[2, [#item_custom_name[3, [#item_custom_name[4, [#item_custom_name[5| Summary:The SNO+ Collaboration reports new results on reactor antineutrino oscillations using data acquired from May 2022 through July 2025. The spectral analysis of a flux dominated by nuclear reactors at 240, 350, and 355 kilometers yields the mass-squared difference $Δm^2_21=(7.93^+0.21_-0.24)times 10^-5$ eV$^2$. […]


Continue.. Measurement of reactor antineutrino oscillations with 1.46 ktonne-years of data at SNO+

Effective Hamiltonians for Ge/Si core/shell nanowires from higher order perturbation theory

Kavli Affiliate: Michael Wimmer | Summary:We theoretically explore the electronic structure of holes in cylindrical Germanium/Silicon core/shell nanowires using a perturbation theory approach. The approach yields a set of interpretable and transferable effective low-energy models for the lowest few sub-bands up to fifth order for experimentally relevant growth directions. In particular, we are able to […]


Continue.. Effective Hamiltonians for Ge/Si core/shell nanowires from higher order perturbation theory

Human-AI collaborative autonomous synthesis with pulsed laser deposition for remote epitaxy

Kavli Affiliate: Daniel C. Ralph| First 5 Authors: Asraful Haque, Asraful Haque, , , | Summary:Autonomous laboratories typically rely on data-driven decision-making, occasionally with human-in-the-loop oversight to inject domain expertise. Fully leveraging AI agents, however, requires tightly coupled, collaborative workflows spanning hypothesis generation, experimental planning, execution, and interpretation. To address this, we develop and deploy […]


Continue.. Human-AI collaborative autonomous synthesis with pulsed laser deposition for remote epitaxy

CountSteer: Steering Attention for Object Counting in Diffusion Models

Kavli Affiliate: Hsiao-Mei (Sherry) Cho| First 5 Authors: [#item_custom_name[1, [#item_custom_name[2, [#item_custom_name[3, [#item_custom_name[4, [#item_custom_name[5| Summary:Text-to-image diffusion models generate realistic and coherent images but often fail to follow numerical instructions in text, revealing a gap between language and visual representation. Interestingly, we found that these models are not entirely blind to numbers-they are implicitly aware of their […]


Continue.. CountSteer: Steering Attention for Object Counting in Diffusion Models

High Mobility Multiple-Channel AlScN/GaN Heterostructures

Kavli Affiliate: Grace Xing | Summary:Aluminum scandium nitride (AlScN) is a promising barrier material for gallium nitride (GaN)-based transistors for the next generation of radio-frequency electronic devices. In this work, we examine the transport properties of two dimensional electron gases (2DEGs) in single- and multi-channel AlScN/GaN heterostructures grown by molecular beam epitaxy, and demonstrate the […]


Continue.. High Mobility Multiple-Channel AlScN/GaN Heterostructures

STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

Kavli Affiliate: Michael Turner| First 5 Authors: Brad Shook, Brad Shook, , , | Summary:Time series foundation models (TSFMs) pretrained on data from multiple domains have shown strong performance on diverse modeling tasks. Various efforts have been made to develop foundation models specific to electroencephalography (EEG) data, which records brain electrical activity as time series. […]


Continue.. STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

Kavli Affiliate: Michael Turner| First 5 Authors: Brad Shook, Brad Shook, , , | Summary:Time series foundation models (TSFMs) pretrained on data from multiple domains have shown strong performance on diverse modeling tasks. Various efforts have been made to develop foundation models specific to electroencephalography (EEG) data, which records brain electrical activity as time series. […]


Continue.. STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

What’s Inside Matters: The Effect of Oxygen Fugacity and Initial Volatile Abundance on the Atmospheres of the TRAPPIST-1 Planets

Kavli Affiliate: Laura Schaefer | Summary:The TRAPPIST-1 planets have become prime targets for studying the atmospheric and geophysical properties of planets around M-dwarf stars. To effectively identify their atmospheric composition, we first must understand their geological evolution. For this study, we focus on enhancing an existing atmosphere-interior exchange model by incorporating additional geological processes relevant […]


Continue.. What’s Inside Matters: The Effect of Oxygen Fugacity and Initial Volatile Abundance on the Atmospheres of the TRAPPIST-1 Planets