Tootfinder

TaxCalcBench: Evaluating Frontier Models on the Tax Calculation Task
Michael R. Bock, Kara Molisee, Zachary Ozer, Sumit Shah
https://arxiv.org/abs/2507.16126

TaxCalcBench: Evaluating Frontier Models on the Tax Calculation Task
Can AI file your taxes? Not yet. Calculating US personal income taxes is a task that requires building an understanding of vast amounts of English text and using that knowledge to carefully compute results. We propose TaxCalcBench, a benchmark for determining models' abilities to calculate personal income tax returns given all of the necessary information. Our experiment shows that state-of-the-art models succeed in calculating less than a third of federal income tax returns even on this simpli…

How? This is disaster prep. Imagine the US is collapsing (because it is). What do you need to do to keep yourself and your community safe when things shut down for 24 hours, 48, 72? What do you do when the government leaves and never comes back? Build that.
This isn't bunker and beans shit either. We all die if we try to live alone. The only way to survive is together. Once you build the thing that keeps you and your friends alive, expand... Go bigger.... Organize with more people.
That's a clear and actionable strategy that beats the hell out of "vote for the lesser of two evils until everything is evil."

GenTune: Toward Traceable Prompts to Improve Controllability of Image Refinement in Environment Design
Wen-Fan Wang, Ting-Ying Lee, Chien-Ting Lu, Che-Wei Hsu, Nil Ponsa Campany, Yu Chen, Mike Y. Chen, Bing-Yu Chen
https://arxiv.org/abs/2508.15227

GenTune: Toward Traceable Prompts to Improve Controllability of Image Refinement in Environment Design
Environment designers in the entertainment industry create imaginative 2D and 3D scenes for games, films, and television, requiring both fine-grained control of specific details and consistent global coherence. Designers have increasingly integrated generative AI into their workflows, often relying on large language models (LLMs) to expand user prompts for text-to-image generation, then iteratively refining those prompts and applying inpainting. However, our formative study with 10 designers su…

QWD-GAN: Quality-aware Wavelet-driven GAN for Unsupervised Medical Microscopy Images Denoising
Qijun Yang, Yating Huang, Lintao Xiang, Hujun Yin
https://arxiv.org/abs/2509.15814

QWD-GAN: Quality-aware Wavelet-driven GAN for Unsupervised Medical Microscopy Images Denoising
Image denoising plays a critical role in biomedical and microscopy imaging, especially when acquiring wide-field fluorescence-stained images. This task faces challenges in multiple fronts, including limitations in image acquisition conditions, complex noise types, algorithm adaptability, and clinical application demands. Although many deep learning-based denoising techniques have demonstrated promising results, further improvements are needed in preserving image details, enhancing algorithmic e…

Lensed stars in galaxy-galaxy strong lensing -- a JWST prediction for the Cosmic Horseshoe
Sung Kei Li, Luke Weisenbach, Thomas E. Collett, Jose M. Diego, Jeremy Lim, Thomas J. Broadhurst, Alex Chow, Wolfgang J. R. Enzi, Patrick L. Kelly, Carlos R. Melo-Carneiro, Jose M. Palencia, Liliya L. R. Williams, Jiashuo Zhang
https://arxiv.org/abs/…

Lensed stars in galaxy-galaxy strong lensing -- a JWST prediction for the Cosmic Horseshoe
We explore for the first time the possibility of detecting lensed star transients in galaxy-galaxy strong lensing systems upon repeated, deep imaging using the {\it James-Webb Space Telescope} ({\it JWST}). Our calculation predicts that the extremely high recent star formation rate of $\sim 100\,M_{\odot}\textrm{yr}^{-1}$ over the last 50 Myr (not accounting for image multiplicity) in the ``Cosmic Horseshoe'' lensed system ($z = 2.381$) generates many young, bright stars, of which their large a…

3I/ATLAS (C/2025 N1): Direct Spacecraft Exploration of a Possible Relic of Planetary Formation at "Cosmic Noon"
T. Marshall Eubanks, Bruce G. Bills, Adam Hibberd, W. Paul Blase, Andreas M. Hein, Robert G. Kennedy III, Adrien Coffinet, Jean Schneider, Pierre Kervella, Carlos Gomez de Olea Ballester
https://arxiv.org/abs/2508.15768…

3I/ATLAS (C/2025 N1): Direct Spacecraft Exploration of a Possible Relic of Planetary Formation at "Cosmic Noon"
The interstellar object 3I/ATLAS (also C/2025 N1 (ATLAS), henceforth, 3I), discovered by the ATLAS Chile telescope on 2025 July 1, was rapidly revealed to be the third known interstellar object (ISO) transiting the solar system, with an incoming velocity at infinity of 57.9763 $\pm$ 0.0044 km s$^{-1}$. An examination of 3I's pre-encounter kinematics shows that it is likely to be an object from the galactic thick disk, and thus a remnant of the Galaxy's ``cosmic noon'' period of intense star for…

Resolving Black Hole Family Issues Among the Massive Ancestors of Very High-Spin Gravitational-Wave Events Like GW231123
Jakob Stegmann, Aleksandra Olejak, Selma E. de Mink
https://arxiv.org/abs/2507.15967

Resolving Black Hole Family Issues Among the Massive Ancestors of Very High-Spin Gravitational-Wave Events Like GW231123
The latest detection of GW231123, a binary black hole (BH) merger with exceptionally large masses and high spins for the incoming components, has been suggested as a smoking gun for hierarchical formation. In this scenario, a first generation of BHs resulting from collapsing stars form in a dense environment. Here they can assemble dynamically and undergo subsequent mergers. We discuss three challenges for the formation of a GW231123-like event inside a star cluster: 1) The high masses of the i…

The GECKOS Survey: revealing the formation history of a barred galaxy via structural decomposition and resolved spectroscopy
A. Fraser-McKelvie, D. A. Gadotti, F. Fragkoudi, C. de S\'a-Freitas, M. Martig, T. Davis, R. Elliott, D. Fisher, M. R. Hayden, J. van de Sande, A. B. Watts
https://arxiv.org/abs/2509.15976

The GECKOS Survey: revealing the formation history of a barred galaxy via structural decomposition and resolved spectroscopy
Disentangling the (co-)evolution of individual galaxy structural components remains a difficult task owing to the inability to cleanly isolate light from spatially overlapping components. In this pilot study of PGC 044931, observed as part of the GECKOS survey, we utilise VIRCAM $H$-band imaging to decompose the galaxy into five photometric components, three of which contribute $>50\%$ of light in given regions: a main disc, a boxy/peanut bulge, and a nuclear disc. When the photometric decompos…

TOM: An Open-Source Tongue Segmentation Method with Multi-Teacher Distillation and Task-Specific Data Augmentation
Jiacheng Xie, Ziyang Zhang, Biplab Poudel, Congyu Guo, Yang Yu, Guanghui An, Xiaoting Tang, Lening Zhao, Chunhui Xu, Dong Xu
https://arxiv.org/abs/2508.14932

TOM: An Open-Source Tongue Segmentation Method with Multi-Teacher Distillation and Task-Specific Data Augmentation
Tongue imaging serves as a valuable diagnostic tool, particularly in Traditional Chinese Medicine (TCM). The quality of tongue surface segmentation significantly affects the accuracy of tongue image classification and subsequent diagnosis in intelligent tongue diagnosis systems. However, existing research on tongue image segmentation faces notable limitations, and there is a lack of robust and user-friendly segmentation tools. This paper proposes a tongue image segmentation model (TOM) based on…

The Fraction of Clumpy Galaxies in JWST Surveys over $2<z<12$
Alexander de la Vega, Bahram Mobasher, Faezeh Manesh, Niloofar Sharei, Nima Chartab, Zahra Sattari
https://arxiv.org/abs/2508.14972

The Fraction of Clumpy Galaxies in JWST Surveys over $2
High-redshift galaxies exhibit compact regions of intense star formation, known as ``clumps,'' which are conspicuous in the rest-frame ultraviolet. Studying them can shed light on how they form and evolve and inform theoretical models of galaxy evolution. We examine the evolution of clumpy galaxies with redshift and stellar mass over $2