PromptTuner: SLO-Aware Elastic System for LLM Prompt Tuning

Kavli Affiliate: Wei Gao| Summary:Prompt tuning has become a prominent strategy for enhancing the performance of Large Language Models (LLMs) on downstream tasks. Many IT enterprises now offer Prompt-Tuning-as-a-Service to fulfill the growing demand for prompt tuning LLMs on downstream tasks. Their primary objective is to satisfy users Service Level Objectives (SLOs) while reducing resource […]


Continue.. PromptTuner: SLO-Aware Elastic System for LLM Prompt Tuning

Single-Particle Resonant States in Relativistic Hartree-Fock Theory: A Green’s Function Approach

Kavli Affiliate: Wei Gao| Summary:Relativistic Hartree-Fock theory is combined with the Green’s function method in coordinate space to study both single-particle bound and resonant states within a unified framework. Within this approach, single-particle resonance energies and widths are unambiguously extracted from the density of states, and the influence of the Coulomb exchange effects on proton […]


Continue.. Single-Particle Resonant States in Relativistic Hartree-Fock Theory: A Green’s Function Approach

vCA1 SST neurons represent avoidance states that guide anxiety-related behavioral choices

Kavli Affiliate: Mazen Kheirbek | Authors: Joshua X Bratsch-Prince, Mingyang Wei, Srutika Sureshbabu, Vivan Chhaya, Nuria Cano Adamuz, Kelsey Logas and Mazen Kheirbek | Summary: In uncertain, threatening environments, rapid and flexible behavioral decisions are essential for survival. The ventral CA1 region of the hippocampus encodes threatening contexts and can regulate avoidance-related behaviors. While activity […]


Continue.. vCA1 SST neurons represent avoidance states that guide anxiety-related behavioral choices

Granule cells reorient cortical manifolds to separate contexts but preserve their geometry

Kavli Affiliate: Rui Costa | Authors: Martha G Garcia-Garcia, Michal Jozef Wojcik, Srijan Thota, Luke Drake, Amma Otchere, Oluwatobi Akinwale, Lizmaylin Ramos, Rui Ponte Costa and Mark J Wagner | Summary: To learn effectively, animals must generalize across yet distinguish between related contexts. Generalization relies on low-dimensional neural manifolds found throughout neocortex1–3, which accelerate learning […]


Continue.. Granule cells reorient cortical manifolds to separate contexts but preserve their geometry

Repeating Nuclear Transients from Repeating Partial Tidal Disruption Events

Kavli Affiliate: Dheeraj Pasham | Summary:Extragalactic nuclear transients that exhibit repeating outbursts can be modeled as the repeated dynamical interaction between bound stars and supermassive black holes (SMBHs). A subset of these transients, with recurrence timescales of months-to-years, have been explained as accretion flares from the repeated tidal stripping of a star by an SMBH, […]


Continue.. Repeating Nuclear Transients from Repeating Partial Tidal Disruption Events

Nonlinear Dynamics in General Relativity

Kavli Affiliate: Ulrich Sperhake | Summary:Black holes and gravitational waves are consequences of the nonlinear character of the Einstein equations. Yet, the remarkable properties of General Relativity point to the existence of other effects. Here we uncover new nonlinear facets of gravity. We establish higher harmonic generation, spectral broadening and focusing in the Einstein Klein-Gordon […]


Continue.. Nonlinear Dynamics in General Relativity

The erasure of Galactic bar resonances by dark matter subhaloes

Kavli Affiliate: Lina Necib | Summary:In the context of increasing appreciation for the coupling between the Galactic bar and the halo, we introduce a new framework using stars trapped in resonance with the bar to probe the Galactic dark matter subhalo population. Since resonant stars occupy a finite width in action space, perturbations from subhaloes […]


Continue.. The erasure of Galactic bar resonances by dark matter subhaloes

Post-inflationary axion constraints from the Lyman-$α$ forest

Kavli Affiliate: Martin Haehnelt | Summary:Among the most compelling cold dark matter candidates, the axion has recently been subject to a wide range of astrophysical studies aiming to constraints its properties. We present updated bounds on the isocurvature fraction, $f_rmiso$, which parameterizes the contribution of isocurvature perturbations induced by post-inflationary produced axion-like particles (ALPs) to […]


Continue.. Post-inflationary axion constraints from the Lyman-$α$ forest

Exploring gas thermodynamics around galaxies from the Sunyaev-Zel’dovich effects: impact of galaxy-halo connection, 2D projection and velocity field

Kavli Affiliate: Risa Wechsler | Summary:A complete picture of the gas thermodynamics around galaxies is imprinted on the cosmic microwave background (CMB). Indeed, the thermal, kinematic, and relativistic Sunyaev-Zel’dovich effects (tSZ, kSZ, rSZ) measure the gas density, temperature, pressure of baryonic feedback and bulk velocity around galaxies, along with the gravitational potential it sits in. […]


Continue.. Exploring gas thermodynamics around galaxies from the Sunyaev-Zel’dovich effects: impact of galaxy-halo connection, 2D projection and velocity field

Semi-Supervised Generative Learning via Latent Space Distribution Matching

Kavli Affiliate: Feng Long| Summary: We introduce Latent Space Distribution Matching (LSDM), a novel framework for semi-supervised generative modeling of conditional distributions. LSDM operates in two stages: (i) learning a low-dimensional latent space from both paired and unpaired data, and (ii) performing joint distribution matching in this space via the 1-Wasserstein distance, using only paired […]


Continue.. Semi-Supervised Generative Learning via Latent Space Distribution Matching