SCALE: Upscaled Continual Learning of Large Language 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:We revisit continual pre-training for large language models and argue that progress now depends more on scaling the right structure than on scaling parameters alone. We introduce SCALE, a width upscaling architecture that inserts lightweight expansion into linear modules while freezing all […]


Continue.. SCALE: Upscaled Continual Learning of Large Language Models

Adobe Summit Concierge Evaluation with Human in the Loop

Kavli Affiliate: Dan Luo | First 5 Authors: Yiru Chen, Yiru Chen, , , | Summary: Generative AI assistants offer significant potential to enhance productivity, streamline information access, and improve user experience in enterprise contexts. In this work, we present Summit Concierge, a domain-specific AI assistant developed for Adobe Summit. The assistant handles a wide […]


Continue.. Adobe Summit Concierge Evaluation with Human in the Loop

The Challenge in Illuminating the Invisible: Constraining LyC Escape with Bayesian Modelling and Symbolic Regression

Kavli Affiliate: Roberto Maiolino | Summary:Direct observations of Lyman continuum (LyC) radiation from galaxies during the Epoch of Reionization (EoR) are impeded by absorption in the intergalactic medium, requiring indirect methods to infer the escape fraction of ionizing photons ($f_rm esc^rm LyC$). One approach is to develop and validate such methods on local analogues of […]


Continue.. The Challenge in Illuminating the Invisible: Constraining LyC Escape with Bayesian Modelling and Symbolic Regression

Solving the cooling flow problem with combined jet-wind AGN feedback

Kavli Affiliate: Feng Yuan | First 5 Authors: Aoyun He, Aoyun He, , , | Summary: Active galactic nucleus (AGN) feedback is widely viewed as the most promising solution to the long-standing cooling flow problem in galaxy clusters, yet previous models prescribe jet properties inconsistent with accretion physics. We perform high-resolution hydrodynamic simulations of a […]


Continue.. Solving the cooling flow problem with combined jet-wind AGN feedback

Dynamical Tidal Response of Schwarzschild Black Holes

Kavli Affiliate: Austin Joyce | Summary:Dynamical Love numbers capture the conservative response of an object to a time-dependent external tidal gravitational field. We compute the dynamical Love numbers of Schwarzschild black holes in general relativity within a point-particle effective field theory framework. In addition to the known logarithmic running, we compute the finite scheme-dependent contributions […]


Continue.. Dynamical Tidal Response of Schwarzschild Black Holes

Dynamical Tidal Response of Schwarzschild Black Holes

Kavli Affiliate: Austin Joyce | Summary:Dynamical Love numbers capture the conservative response of an object to a time-dependent external tidal gravitational field. We compute the dynamical Love numbers of Schwarzschild black holes in general relativity within a point-particle effective field theory framework. In addition to the known logarithmic running, we compute the finite scheme-dependent contributions […]


Continue.. Dynamical Tidal Response of Schwarzschild Black Holes

Dynamical Tidal Response of Schwarzschild Black Holes

Kavli Affiliate: Austin Joyce| First 5 Authors: Oscar Combaluzier–Szteinsznaider, Oscar Combaluzier–Szteinsznaider, , , | Summary:Dynamical Love numbers capture the conservative response of an object to a time-dependent external tidal gravitational field. We compute the dynamical Love numbers of Schwarzschild black holes in general relativity within a point-particle effective field theory framework. In addition to the […]


Continue.. Dynamical Tidal Response of Schwarzschild Black Holes

Search for Diffuse Supernova Neutrino Background with 956.2 days of Super-Kamiokande Gadolinium Dataset

Kavli Affiliate: Masayuki Nakahata| First 5 Authors: K. Abe, K. Abe, , , | Summary:We report the search result for the Diffuse Supernova Neutrino Background (DSNB) in neutrino energies beyond 9.3~MeV in the gadolinium-loaded Super-Kamiokande (SK) detector with $22,500times956.2$$~rm m^3cdot day$ exposure. %$22.5rm ktimes956.2$$~rm m^3cdot day$ exposure. Starting in the summer of 2020, SK introduced […]


Continue.. Search for Diffuse Supernova Neutrino Background with 956.2 days of Super-Kamiokande Gadolinium Dataset