Zobrazeno 1 - 10
of 3 587
pro vyhledávání: '"Asthana P"'
Evaluating large vision-language models (LVLMs) is very expensive, due to the high computational costs and the wide variety of tasks. The good news is that if we already have some observed performance scores, we may be able to infer unknown ones. In
Externí odkaz:
http://arxiv.org/abs/2410.10112
Autor:
Zhang, Bing, Takeuchi, Mikio, Kawahara, Ryo, Asthana, Shubhi, Hossain, Md. Maruf, Ren, Guang-Jie, Soule, Kate, Zhu, Yada
The advancement of large language models (LLMs) has led to a greater challenge of having a rigorous and systematic evaluation of complex tasks performed, especially in enterprise applications. Therefore, LLMs need to be able to benchmark enterprise d
Externí odkaz:
http://arxiv.org/abs/2410.12857
Autor:
Asthana, Shikhar, Haehnelt, Martin G., Kulkarni, Girish, Bolton, James S., Gaikwad, Prakash, Keating, Laura C., Puchwein, Ewald
The relative contribution of emission from stellar sources and accretion onto supermassive black holes to reionization has been brought into focus again by the apparent high abundance of faint AGN at $4\lesssim z\lesssim11$ uncovered by JWST. We inve
Externí odkaz:
http://arxiv.org/abs/2409.15453
We propose in this paper an analytically new construct of a diffusion model whose drift and diffusion parameters yield an exponentially time-decaying Signal to Noise Ratio in the forward process. In reverse, the construct cleverly carries out the lea
Externí odkaz:
http://arxiv.org/abs/2408.08306
Autor:
Asthana, Shikhar, Haehnelt, Martin G., Kulkarni, Girish, Aubert, Dominique, Bolton, James S., Keating, Laura C.
Publikováno v:
MNRAS 533, 2024, 2843-2866
We present a new suite of late-end reionization simulations performed with ATON-HE, a revised version of the GPU-based radiative transfer code ATON that includes helium. The simulations are able to reproduce the Ly$\alpha$ flux distribution of the E-
Externí odkaz:
http://arxiv.org/abs/2404.06548
Large vision-language models (LVLMs), designed to interpret and respond to human instructions, occasionally generate hallucinated or harmful content due to inappropriate instructions. This study uses linear probing to shed light on the hidden knowled
Externí odkaz:
http://arxiv.org/abs/2403.09037
Neural architecture search automates the design of neural network architectures usually by exploring a large and thus complex architecture search space. To advance the architecture search, we present a graph diffusion-based NAS approach that uses dis
Externí odkaz:
http://arxiv.org/abs/2403.06020
Feature shaping refers to a family of methods that exhibit state-of-the-art performance for out-of-distribution (OOD) detection. These approaches manipulate the feature representation, typically from the penultimate layer of a pre-trained deep learni
Externí odkaz:
http://arxiv.org/abs/2402.00865
Autor:
Gupta, Kartik, Asthana, Akshay
Quantized networks use less computational and memory resources and are suitable for deployment on edge devices. While quantization-aware training QAT is the well-studied approach to quantize the networks at low precision, most research focuses on ove
Externí odkaz:
http://arxiv.org/abs/2311.05109
Autor:
Asthana, Anant, Goyal, Shreev
Let $G$ be a connected multigraph with $n$ vertices, and suppose $G$ has been edge-colored with $n-1$ colors so that each color class induces a spanning tree. Rota's Basis Conjecture for graphic matroids posits that one can find $n-1$ mutually edge-d
Externí odkaz:
http://arxiv.org/abs/2310.19242