Tootfinder

Mass killings
Was looking through Wikipedia's list of mass killings in America (#guns #GunViolence #Shooting

3D Aware Region Prompted Vision Language Model
An-Chieh Cheng, Yang Fu, Yukang Chen, Zhijian Liu, Xiaolong Li, Subhashree Radhakrishnan, Song Han, Yao Lu, Jan Kautz, Pavlo Molchanov, Hongxu Yin, Xiaolong Wang, Sifei Liu
https://arxiv.org/abs/2509.13317

3D Aware Region Prompted Vision Language Model
We present Spatial Region 3D (SR-3D) aware vision-language model that connects single-view 2D images and multi-view 3D data through a shared visual token space. SR-3D supports flexible region prompting, allowing users to annotate regions with bounding boxes, segmentation masks on any frame, or directly in 3D, without the need for exhaustive multi-frame labeling. We achieve this by enriching 2D visual features with 3D positional embeddings, which allows the 3D model to draw upon strong 2D priors…

Finding a solution to the Erd\H{o}s-Ginzburg-Ziv theorem in $O(n\log\log\log n)$ time
Yui Hin Arvin Leung
https://arxiv.org/abs/2507.08139 https://

Finding a solution to the Erdős-Ginzburg-Ziv theorem in $O(n\log\log\log n)$ time
The Erdős-Ginzburg-Ziv theorem states that for any sequence of $2n-1$ integers, there exists a subsequence of $n$ elements whose sum is divisible by $n$. In this article, we provide a simple, practical $O(n\log\log n)$ algorithm and a theoretical $O(n\log\log\log n)$ algorithm, both of which improve upon the best previously known $O(n\log n)$ approach. This shows that a specific variant of boolean convolution can be implemented in time faster than the usual $O(n\log n)$ expected from FFT-based…

InfGen: A Resolution-Agnostic Paradigm for Scalable Image Synthesis
Tao Han, Wanghan Xu, Junchao Gong, Xiaoyu Yue, Song Guo, Luping Zhou, Lei Bai
https://arxiv.org/abs/2509.10441

InfGen: A Resolution-Agnostic Paradigm for Scalable Image Synthesis
Arbitrary resolution image generation provides a consistent visual experience across devices, having extensive applications for producers and consumers. Current diffusion models increase computational demand quadratically with resolution, causing 4K image generation delays over 100 seconds. To solve this, we explore the second generation upon the latent diffusion models, where the fixed latent generated by diffusion models is regarded as the content representation and we propose to decode arbit…

Refugees, intergenerational trauma, child death, abusive family
Also just finished "The Best We Could Do" by Thi Bui, which is the second memoir I've stumbled upon recently that deals with the Vietnamese exodus after the end of the war (House Without Walls by Ching Yeung Russel is the other one, which is written in verse, not illustrated). Bui traces more of the political landscape and history of Vietnam through the stories of both of her parents, and also unpacks a lot of intergenerational trauma, but has less focus on the boat trip out and refugee camp experience, presumably because hers were easier than Russel's.
My thoughts after reading this return repeatedly to all of the impacts that patriarchy and toxic masculinity had on her father, from setting up his father and grandfather to be abusive towards him and the women in their lives, to pushing him deep into depression when he feels unable to fulfill the role of a protective husband, ironically leaving his wife to pick up the slack and ultimately ruining their relationship, to how it teaches him to despise and shirk the caregiver role he's left with, ultimately passing on some measure of trauma to his children. For sure war, abusive family, and child death can happen in the absence of patriarchy and those are in some ways perhaps bigger factors here, but at the same time, Bui's mom copes with most of the same factors in healthier ways.
#AmReading

This band Ultra Lights are often described as "90s slacker rock". To my ears, upon hearing this, I thought it was a new Soft Pack (aka Muslims) album lol
https://www.ultralightsband.com/

Ultra Lights
Rainbow Vision is a planner brand and creative consulting practice by Leela Hoehn. After 10+ years of building and expanding the Rainbow Vision paper and gift brand Leela condensed the line down to a handful of favorites in 2023. Rainbow Vision is based out of Atlanta, GA

Microstructure of Silicon Anodes in Solid-State Batteries -- From Crystalline to Amorphous
Shamail Ahmed, Federico Rossi, Hanyu Huo, Johannes Haust, Franziska Hueppe, Juergen Belz, Andreas Beyer, Juergen Janek, Kerstin Volz
https://arxiv.org/abs/2507.16561

Microstructure of Silicon Anodes in Solid-State Batteries -- From Crystalline to Amorphous
Silicon offers great promise as a potential anode active material and the optimum alternative to lithium metal in all-solid-state lithium-ion batteries. However, its practical application is limited by severe volume expansion (~300%) during lithiation, leading to cracking upon delithiation. In this study, we investigated the microstructural evolution of microcrystalline silicon electrodes in a solid-electrolyte-free environment using cryogenic scanning transmission electron microscopy (STEM) du…

Efficient and Generalizable Speaker Diarization via Structured Pruning of Self-Supervised Models
Jiangyu Han, Petr P\'alka, Marc Delcroix, Federico Landini, Johan Rohdin, Jan Cernock\'y, Luk\'a\v{s} Burget
https://arxiv.org/abs/2506.18623

Efficient and Generalizable Speaker Diarization via Structured Pruning of Self-Supervised Models
Self-supervised learning (SSL) models such as WavLM have brought substantial improvements to speaker diarization by providing rich contextual representations. However, the high computational and memory costs of these models hinder their deployment in real-time and resource-constrained scenarios. In this work, we present a comprehensive study on compressing SSL-based diarization models through structured pruning guided by knowledge distillation. Building upon our previous work, we extend the ana…

A Multi-Stage Fine-Tuning and Ensembling Strategy for Pancreatic Tumor Segmentation in Diagnostic and Therapeutic MRI
Omer Faruk Durugol, Maximilian Rokuss, Yannick Kirchhoff, Klaus H. Maier-Hein
https://arxiv.org/abs/2508.21775

A Multi-Stage Fine-Tuning and Ensembling Strategy for Pancreatic Tumor Segmentation in Diagnostic and Therapeutic MRI
Automated segmentation of Pancreatic Ductal Adenocarcinoma (PDAC) from MRI is critical for clinical workflows but is hindered by poor tumor-tissue contrast and a scarcity of annotated data. This paper details our submission to the PANTHER challenge, addressing both diagnostic T1-weighted (Task 1) and therapeutic T2-weighted (Task 2) segmentation. Our approach is built upon the nnU-Net framework and leverages a deep, multi-stage cascaded pre-training strategy, starting from a general anatomical …