Tootfinder

Managed to repair my digital piano. Two of the c's stopped working. Turns out the fix was easy: I used some 9B pencil to rub the connectors on the piano, as they apparently work with graphite. There was a silicone bit I could take off to get to the connector.
Opening the piano up was by far the hardest part of this repair.
#RightToRepair

It is simply unbelievable how #enshittification has ruined all of Docker’s websites and is still making them worse and worse. As a part-time user of Docker’s #OpenSource components, usually doing what I think should be regarded most basic and elementary stuff, I cannot find the few snipp…

If you don't like Debian, maybe rocks just aren't your thing, I love rocks, they sit there, never move, never break, just stable, just like Debian. :debian:
#Linux #Debian #Rock

Sona Bashirova steht als Deutsche Meisterin der U10w fest! Herzlichen Glückwunsch 🎉 #dem25

#Hamas City.

SCOTUS rejects NetChoice's emergency appeal to block a Mississippi law that requires age verification and parental consent for teens to join social networks (New York Times)
https://www.nytimes.com/2025/08/14/us…

Supreme Court Allows Mississippi Law on Children’s Use of Social Media, for Now
A trade group representing sites like Facebook and X said the law ran afoul of the First Amendment.

@… I can relate. (and some of it is in my closet)

The party whose base believes there's a secret network of politically-connected child-abusers... should probably avoid these headlines.

Full deployment.

Gannett is launching a morning newscast, Connecting America, on its USA Today streaming channel and on YouTube, on August 18; it will air weekdays 7am-9am ET (Matthew Keys/TheDesk.net)
https://thedesk.net/2025/08/usa-today-morning-show-connecting-america/

USA Today to launch morning newscast on streaming channel
A new morning show called "Connecting America" will launch on USA Today's FAST channel, the company announced on Thursday.

ego_social: Ego networks in social media (2012)
Ego networks associated with a set of accounts of three social media platforms (Facebook, Google , and Twitter). Datasets include node features (profile metadata), circles, and ego networks, and were crawled from public sources in 2012.
This network has 168 nodes and 892 edges.
Tags: Social, Online, Unweighted, Metadata

Die Bundesregierung will Cyberschutz für kritische Infrastruktur bis 2026 zur Pflicht machen – und hinkt damit der EU-Richtlinie und der Realität hinterher. Während Angriffe zunehmen, fehlt es an Tempo und Klarheit. Claudia Plattner (BSI) mahnt: Viele Unternehmen haben die Gefahr noch immer nicht auf dem Schirm #ZEIT
#BSI

BSI: Cyberschutz für kritische Infrastruktur soll 2026 zur Pflicht werden
Für rund 29.000 wichtige Unternehmen soll es bald verbindliche Cybersicherheitsvorgaben geben. Die Bundesregierung will eine EU-Richtlinie dazu gesetzlich verankern.

Pedazo de #granizada que ha caído en Villalba. Hielo como aceitunas grandes. Qué susto se ha pegado la perra, porque nos ha pillado en la calle

Kuro muß grade hinter Schränken unterwegs gewesen sein

HereWeAre
Great Australian Pods Podcast Directory: #GreatAusPods

Whaaaat? Swarm hat gerelauncht und funktioniert immer noch wie früher?! Da werden Erinnerungen wach.

#pastpuzzle 96
🟩🟥🟩🟥 ( 501)
🟩🟥🟥🟥 ( 168)
🟩🟩🟥🟥 ( 23)
🟩🟩🟩🟩 (0)
4/4 🟩
https://www.pastpuzzle.de

Bluesky revises its policies and Community Guidelines to comply with new regulations, including EU's DSA, the UK Online Safety Act, and the US Take It Down Act (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/08/14/bluesky-roll…

Bluesky rolls out massive revamp to policies and Community Guidelines | TechCrunch
Some of the changes represent an effort by Bluesky to purposefully shape its community and the behavior of its users, nudging them to be nicer and more respectful of others.

So scene kit is officially deprecated now - that makes me sad!

#DavidHasselhoff #AlexandraPaul
Baywatch Nights 2x19 - The Eighth Seal
#RandomBaywatch #lvdlpx

Atmospheric carbon dioxide levels break seasonal records: NOAA https://ground.news/article/annual-carbon-dioxide-peak-passes-another-milestone_e224ea

Old servers are prepped and ready for their new owner and new home.
#homelab

📰 Hier findest Du täglich Informationen zu #Klima & #Wissenschaft auf Deutsch. Mit Deiner Unterstützung kann ich das weiterhin kostenlose tun. Ich möchte helfen, #Sprachbarrieren zu überb…

KlimaCrew
Ich informiere über Zukunftsthemen, um Menschen vor Risiken zu warnen und Fortschritte sichtbar zu machen.

"Localhost tracking" explicado https://www.zeropartydata.es/p/localhost-tracking-explained-it-could

New article,
Degrowth as an Essential Part of an Eco-Socialist Transition
by Anna Gregoletto, on DegrowthUK
https://degrowthuk.org/2025/06/12/degrowth-as-an-essential-part-of-an-eco-socialist-transition/

Degrowth as an Essential Part of an Eco-Socialist Transition
by Anna Gregoletto* In the series Prospects for Degrowth Many thanks to Mark Burton for encouraging me to write this article and offering so many helpful suggestions. Thank you also to my comrades …

Ebe!!

Ebe!!

Activists in NYC demand financial institutions stop funding fossil fuel projects and support renewable energy. #climatechange #climatesolutions #climate

A Week of Gulf South Solidarity in New York City - Inside Climate News
Gulf Coast residents brought their fight against fossil fuel finance to corporate headquarters, connecting local pollution battles with global solidarity movements.

Berlin-based Payrails, whose platform helps enterprises accept payments, issue payouts, and more, raised a $32M Series A, taking its total funding to $52.8M (Tamara Djurickovic/Tech.eu)
https://tech.eu/2025/06/12/payrails-ra

Payrails raises $32M Series A to boost payment performance for global enterprises
A global payment software company Payrails, has raised $32 million in Series A funding to further its mission of enabling enterprises to take control of their payment operations.  Founded by former fintech leaders from Delivery Hero, Payrails’ founding team brings deep experience building global, scalable financial infrastructure. With the continued backing of leading technology investors, Payrails is strengthening its position as a category-defining payment platform for large enterprises. …

The Floorplan
Great Australian Pods Podcast Directory: #GreatAusPods

Wir brauchen alle mehr Ambiguitätstoleranz. Also die Fähigkeit, auch mit Abstufungen und Widersprüchen leben zu können. Nachdem das aber so ein Unwort ist, schlage ich vor, dass wir diese Eigenschaft einfach #ambi nennen.

A group of merging galaxies falling onto Abell 2142
Aashiya Anitha Shaji, Ana\"elle Hall\'e, Damir Gasymov, Anne-Laure Melchior, Fran\c{c}oise Combes, Andrea Cattaneo
https://arxiv.org/abs/2507.08172

A group of merging galaxies falling onto Abell 2142
Galaxy clusters produce a very hostile environment to galaxies, whose gas gets stripped by ram-pressure, suffer galaxy interactions and witness quenching of their star formation. Clusters, like Abell 2142, grow not only through galaxy accretion but also through group infall. Our goal is to study the physical and dynamical state of the most conspicuous infalling group, on a filament projected at 1.3 Mpc from the Abell 2142 centre. The galaxy group is the leading edge of a spectacular trailing 70…

"Zur Sicherheit senden wir Ihnen Zugangsdaten und Aktivierungscode in zwei getrennten Umschlägen."
Ähm ja... Die Umschläge lagen beide am selben Tag im Briefkasten, mit gleichem Poststempel (OK, einer A-Post, einer B-Post) und sind schön auffällig mit *PERSÖNLICH* angeschrieben. 🤦‍♀️
#opsecfailures

Do ants even have toes? 🐜

Fast Ramanujan--type Series for Logarithms. Part II
Jorge Zuniga
https://arxiv.org/abs/2506.10321 https://arxiv.org/pdf/2506.10321

Fast Ramanujan--type Series for Logarithms. Part II
This work extends the results of the preprint Ramanujan type Series for Logarithms, Part I, arXiv:2506.08245, which introduced single hypergeometric type identities for the efficient computing of $\log(p)$, where $p\in\mathbb{Z}_{>1}$. We present novel formulas for arctangents and methods for a very fast multiseries evaluation of logarithms. Building upon a $\mathcal{O}((p-1)^{6})$ Ramanujan type series asymptotic approximation for $\log(p)$ as $p\rightarrow1$, formulas for computing $n$ simult…

@… Feliz rotación nueva!!

Time For 9 o'clock #HashTagGames hosted by @… Let's play!
#RebrandWebsitesOrTech

Heute vor 69 Jahren: Am 11. Juni 1956 kam es zum #Atomtest "Flathead". Die Operation #Redwing war eine US-Serie von 17 Atomtestdetonationen von Mai bis Juli 1956. Sie wurden von der Joint Task Force 7 (JTF7) auf den Atollen

TFW someone randomly drops your browser of choice from #nixpkgs #nixos #vivaldi

that moment when you discover the non-CS specialization of your fellow #RSEng you have known for years, by pure accident / #serendipity
@…

Emergence of distinct relaxation behaviour and Quantum Regression Theorem in the Ultra-strong Coupling Limit
Sakil Khan, Bijay Kumar Agarwalla
https://arxiv.org/abs/2508.08378 h…

Emergence of distinct relaxation behaviour and Quantum Regression Theorem in the Ultra-strong Coupling Limit
In the framework of open quantum systems, we derive the dynamical equation governing two-time correlation functions in the ultra-strong coupling (USC) regime between the system and its environment. Unlike the case of the standard weak-coupling regime, in the USC case, we find distinct relaxation behavior for two-time correlators depending on the types of the operators involved in the correlation function. Interestingly, the Quantum Regression Theorem (QRT) emerges after the fastest relaxation t…

Black hole/quantum machine learning correspondence
Jae-Weon Lee, Zae Young Kim
https://arxiv.org/abs/2506.09678 https://arxiv.org/pdf…

Black hole/quantum machine learning correspondence
We explore a potential connection between the black hole information paradox and the double descent phenomenon in quantum machine learning. Information retrieval from Hawking radiation can be viewed through the lens of quantum linear regression over black hole microstates, with the Page time corresponding to the interpolation threshold, beyond which test error decreases despite overparameterization. Using the Marchenko-Pastur law, we derive the variance in test error for the quantum linear regr…

Good morning from Harlem
#Photography #GoodMorning #Sunrise #NYC

aktuelles #Wetter #Behrendorf #Wische 12-07-2025 06:05
Temperatur: (min: 14,7C) 15,3C (max: 17,0C)
Feuchte: 99%
Luftdruck: 1010,1 hPa (langsam fallend)
Windrichtung: 289 Grad
W…

Another step toward generating “live action” movies and shows.
AMC Networks Is Teaming up with AI Company Runway https://ground.news/article/amc-networks-inks-deal-with-ai-company-runway

The office last night. Palais Royale, South Bend IN.
#music #guitarist

“If you think this doesn’t happen in nature you need to grow up” Via https://bsky.app/profile/brendelbored.bsky.social/post/3lqljqpxmbk2f

SV3.3B: A Sports Video Understanding Model for Action Recognition
Sai Varun Kodathala, Yashwanth Reddy Vutukoori, Rakesh Vunnam
https://arxiv.org/abs/2507.17844 https://

SV3.3B: A Sports Video Understanding Model for Action Recognition
This paper addresses the challenge of automated sports video analysis, which has traditionally been limited by computationally intensive models requiring server-side processing and lacking fine-grained understanding of athletic movements. Current approaches struggle to capture the nuanced biomechanical transitions essential for meaningful sports analysis, often missing critical phases like preparation, execution, and follow-through that occur within seconds. To address these limitations, we int…

TIL that Linux hosts files have supported host names listed multiple times with different ips for each line.
Example:
Server1 192.168.1.10
Server1 192.168.1.11
……..
Last I recalled (early 2000s maybe?) this wasn’t a thing. Yay for new knowledge!

Der lokale Konsum reduziert nicht signifikant den CO2-Fußabdruck der Lebensmittel, da die größten Emissionen aus der Produktion und nicht aus dem Transport stammen. Welche Lebensmittel und Systeme sind am umweltfreundlichsten? Die Antwort überrascht.
https://displayeurope.eu/preview/eay4q

Degrowth as an Essential Part of an Eco-Socialist Transition
by Anna Gregoletto* In the series Prospects for Degrowth Many thanks to Mark Burton for encouraging me to write this article and offering so many helpful suggestions. Thank you also to my comrades at Climate Vanguard. Without their teachings I’d have never been able to articulate the ideas in this article. Do check out Climate Vanguard’s incredible work.

Degrowth as an Essential Part of an Eco-Socialist Transition
by Anna Gregoletto* In the series Prospects for Degrowth Many thanks to Mark Burton for encouraging me to write this article and offering so many helpful suggestions. Thank you also to my comrades …

@… según leo por ahí, es un “joven meteorólogo que usa el sistema de las cabañuelas para hacer sus predicciones”.
Las cabañuelas (no he oído esto en mi vida, Hulio) son, según Wikipedia, “un conjunto de métodos tradicionales de predicción meteorológica a largo plazo sin base científica utilizado en el…

Run Local LLMs on Hardware from $50 to $50,000 - We Test and Compare! https://youtu.be/mUGsv_IHT-g?si=SsLWOdpLdHYrBFEH

Bought a nothing phone to explore Android a bit more. the hardware is super nice. Struggling a bit to cope with the ui things tho. Specially miss pull down to search which i use all the time on iOS side. Also in general things feel a bit more weird and glitchy. Sure I'll get used to it tho. Already installed nix-on-droid which is super nice.

And I just rewatched #TheHatefulEight this week. RIP best actor for roles of total psychopaths in #Tarantino movies, “Joe Gage”—erm, #MichaelMadsen
🙇😢

OpenAI's failed Windsurf deal signals broader M&A challenges for OpenAI until it renegotiates its deal with Microsoft (M.G. Siegler/Spyglass)
https://spyglass.org/openai-windsurf-google/

An OpenAI Acquisition Turns Into a Google 'Hackqusition'...
A rough few weeks may signal some real talent trouble for OpenAI – at the worst possible time

📰 Mit Deiner #Spende förderst Du die kostenlose #Kommunikation wichtiger #Zukunftsthemen: Warnungen zu

KlimaCrew
Ich informiere über Zukunftsthemen, um Menschen vor Risiken zu warnen und Fortschritte sichtbar zu machen.

Good morning from Harlem
#Photography #GoodMorning #Sunrise #NYC

Beyond endoscopy for $\mathsf{GL}_2$ over $\mathbb{Q}$ with ramification 3: contribution of the elliptic part
Yuhao Cheng
https://arxiv.org/abs/2508.07167 https://

Beyond endoscopy for $\mathsf{GL}_2$ over $\mathbb{Q}$ with ramification 3: contribution of the elliptic part
We continue to work on \emph{Beyond Endoscopy} for $\mathsf{GL}_2$ over $\mathbb{Q}$ with ramification at $S = \{\infty, q_1, \dots, q_r\}$ (where $2 \in S$), generalizing the final step of Altuğ's work in the unramified setting. We derive an explicit asymptotic formula for the elliptic part when summing over $n

Enjoing a few days with the GNU TALER team at a castle in Switzerland. The wallet can carry real electronic cash and even used for offline transactions. Technology is ready to be used, not based on blockchain and really free software.Pilot regions are now being onboarded. I moderated a branding and an #ERPnext demo session and submitted a proposal for @…

Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling
Yi Xin, Juncheng Yan, Qi Qin, Zhen Li, Dongyang Liu, Shicheng Li, Victor Shea-Jay Huang, Yupeng Zhou, Renrui Zhang, Le Zhuo, Tiancheng Han, Xiaoqing Sun, Siqi Luo, Mengmeng Wang, Bin Fu, Yuewen Cao, Hongsheng Li, Guangtao Zhai, Xiaohong Liu, Yu Qiao, Peng Gao
https://arxiv.org/abs/…

Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling
We present Lumina-mGPT 2.0, a stand-alone, decoder-only autoregressive model that revisits and revitalizes the autoregressive paradigm for high-quality image generation and beyond. Unlike existing approaches that rely on pretrained components or hybrid architectures, Lumina-mGPT 2.0 is trained entirely from scratch, enabling unrestricted architectural design and licensing freedom. It achieves generation quality on par with state-of-the-art diffusion models such as DALL-E 3 and SANA, while prese…

TIL that Linux hosts files have supported host names listed multiple times with different ips for each line.
Example:
Server1 192.168.1.10
Server1 192.168.1.11
……..
Last I recalled (early 2000s maybe?) this wasn’t a thing. Yay for new knowledge!

The administration is terminating leases for 11 OSHA offices, including the one in Houston. Trump has cut 10 percent of its workforce, and more layoffs could be coming. See Will Texas Workers Ever Get a Break from the Summer Heat? https://ground.news/article/will-texas-workers-ever-get-a-break-from-the-summer-heat

Good morning from Harlem
#Photography #GoodMorning #Sunrise #NYC

📰 Einem US-amerikanischen Konzern Geld aus der Tasche zu leiern, um damit Hosting-Rechnungen für Gratis-Angebote zu bezahlen, kann man machen. Aber schön ist es nicht. Ich würde meine Amazon-Partnerlinks gern loswerden.
Bitte unterstütze meine Arbeit mit einer Steady-Mitgliedschaft: https://steady.page/klimacrew

KlimaCrew
Ich informiere über Zukunftsthemen, um Menschen vor Risiken zu warnen und Fortschritte sichtbar zu machen.

I thought that “Alligator Alcatraz” was a nickname set by oppositors, not the actual name of the disgraceful place.
These people don’t cease to surprise me with their cruelty. https://toad.social/@wdlindsy/114834996595530816

Nadelstecher sind unterwegs. Gestern wurde einer sus dem RailJet in dem ich saß von der #Polizei entfernt. Was motiviert diese Menschen? https://displayeurope.eu/preview/eay4nsvyj2yoargi7lcw…

'The Prime Minister told MPs that “we can’t just tax your way to growth” '
No, you tax the rich because you need to redistribute given that 'growth' is both broken and ecologically lethal.
Starmer rallies in support of the rich as he comes under pressure to tax wealth | Morning Star

How old am I? I'm this old #c64 #ik

[2025-07-25 Fri (UTC), 90 new articles found for cs.CV Computer Vision and Pattern Recognition]
toXiv_bot_toot

„Spielen Sie hier jetzt regelmäßig?“
ich: hmm, ja, also, naja, ich habe mitbekommen, dass diese Gemeinde ab und zu Aushilfen braucht …
ich, innerlich: omg, das ist das allerbeste Kompliment, das Sie jemandem aus der #Kirchenmusik machen können; Sie haben mir gerade mindestens die nächsten 10 Tage versüßt 🥰

Good morning from Harlem
#Photography #GoodMorning #Sunrise #NYC

@… en unas semanas, pero no te veo mal!!

spotify_player has to be one of my favorite tui apps of all time. it's just so very well done https://github.com/aome510/spotify-player

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[5/5]:
- A Survey of Deep Learning for Geometry Problem Solving
Jianzhe Ma, Wenxuan Wang, Qin Jin

@… is always interested in the challenges of #linux, how to package software and how much energy it consumes.

Arago, which is developing a photonic chip, codenamed JEF, that is designed to cut AI energy use by ten times at a comparable cost, raised a $26M seed (Tamara Djurickovic/Tech.eu)
https://tech.eu/2025/07/08/arago-raises-2…

Arago raises $26M to slash AI energy consumption with breakthrough photonic chip
Arago, a deeptech startup pioneering a new class of energy-efficient AI chips powered by light, has raised $26 million in seed funding to accelerate the commercialisation of its photonic processor, codenamed “JEF”. At the core of “JEF” is a proprietary photonic technology that processes data using lasers instead of transistors. By using photons, light particles that generate much less heat than electrons, this chip can deliver the same performance as top GPUs while consuming 10× less …

Wrote a post on creating fully declarative grafana kiosks using #nixos https://marcus.means.no/post/2025-05-25-building-a-grafana-kiosk-with-nixos/

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[4/5]:
- PARTE: Part-Guided Texturing for 3D Human Reconstruction from a Single Image
Hyeongjin Nam, Donghwan Kim, Gyeongsik Moon, Kyoung Mu Lee

Für mich bitte einmal ≪1,5° statt 5%.

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[3/5]:
- Flash-VStream: Efficient Real-Time Understanding for Long Video Streams
Haoji Zhang, Yiqin Wang, Yansong Tang, Yong Liu, Jiashi Feng, Xiaojie Jin

@… ostras! es la primera vez que veo escrito “védel” :))
Ya no se lleva, con los esquís modernos, pero los viejos seguimos haciéndolo de vez en cuando 😅

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[2/5]:
- External Knowledge Injection for CLIP-Based Class-Incremental Learning
Da-Wei Zhou, Kai-Wen Li, Jingyi Ning, Han-Jia Ye, Lijun Zhang, De-Chuan Zhan

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[1/5]:
- ODES: Domain Adaptation with Expert Guidance for Online Medical Image Segmentation
Md Shazid Islam, Sayak Nag, Arindam Dutta, Miraj Ahmed, Fahim Faisal Niloy, Amit K. Roy-Chowdhury

Captain Cinema: Towards Short Movie Generation
Junfei Xiao, Ceyuan Yang, Lvmin Zhang, Shengqu Cai, Yang Zhao, Yuwei Guo, Gordon Wetzstein, Maneesh Agrawala, Alan Yuille, Lu Jiang
https://arxiv.org/abs/2507.18634

Captain Cinema: Towards Short Movie Generation
We present Captain Cinema, a generation framework for short movie generation. Given a detailed textual description of a movie storyline, our approach firstly generates a sequence of keyframes that outline the entire narrative, which ensures long-range coherence in both the storyline and visual appearance (e.g., scenes and characters). We refer to this step as top-down keyframe planning. These keyframes then serve as conditioning signals for a video synthesis model, which supports long context l…

Identifying Prompted Artist Names from Generated Images
Grace Su, Sheng-Yu Wang, Aaron Hertzmann, Eli Shechtman, Jun-Yan Zhu, Richard Zhang
https://arxiv.org/abs/2507.18633 http…

Identifying Prompted Artist Names from Generated Images
A common and controversial use of text-to-image models is to generate pictures by explicitly naming artists, such as "in the style of Greg Rutkowski". We introduce a benchmark for prompted-artist recognition: predicting which artist names were invoked in the prompt from the image alone. The dataset contains 1.95M images covering 110 artists and spans four generalization settings: held-out artists, increasing prompt complexity, multiple-artist prompts, and different text-to-image models. We eval…

SIDA: Synthetic Image Driven Zero-shot Domain Adaptation
Ye-Chan Kim, SeungJu Cha, Si-Woo Kim, Taewhan Kim, Dong-Jin Kim
https://arxiv.org/abs/2507.18632 https://

SIDA: Synthetic Image Driven Zero-shot Domain Adaptation
Zero-shot domain adaptation is a method for adapting a model to a target domain without utilizing target domain image data. To enable adaptation without target images, existing studies utilize CLIP's embedding space and text description to simulate target-like style features. Despite the previous achievements in zero-shot domain adaptation, we observe that these text-driven methods struggle to capture complex real-world variations and significantly increase adaptation time due to their alignmen…

3D Software Synthesis Guided by Constraint-Expressive Intermediate Representation
Shuqing Li, Anson Y. Lam, Yun Peng, Wenxuan Wang, Michael R. Lyu
https://arxiv.org/abs/2507.18625

3D Software Synthesis Guided by Constraint-Expressive Intermediate Representation
Graphical user interface (UI) software has undergone a fundamental transformation from traditional two-dimensional (2D) desktop/web/mobile interfaces to spatial three-dimensional (3D) environments. While existing work has made remarkable success in automated 2D software generation, such as HTML/CSS and mobile app interface code synthesis, the generation of 3D software still remains under-explored. Current methods for 3D software generation usually generate the 3D environments as a whole and can…

SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning
Si-Woo Kim, MinJu Jeon, Ye-Chan Kim, Soeun Lee, Taewhan Kim, Dong-Jin Kim
https://arxiv.org/abs/2507.18616

SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning
Zero-shot Image Captioning (ZIC) increasingly utilizes synthetic datasets generated by text-to-image (T2I) models to mitigate the need for costly manual annotation. However, these T2I models often produce images that exhibit semantic misalignments with their corresponding input captions (e.g., missing objects, incorrect attributes), resulting in noisy synthetic image-caption pairs that can hinder model training. Existing dataset pruning techniques are largely designed for removing noisy text in…

DRWKV: Focusing on Object Edges for Low-Light Image Enhancement
Xuecheng Bai, Yuxiang Wang, Boyu Hu, Qinyuan Jie, Chuanzhi Xu, Hongru Xiao, Kechen Li, Vera Chung
https://arxiv.org/abs/2507.18594

DRWKV: Focusing on Object Edges for Low-Light Image Enhancement
Low-light image enhancement remains a challenging task, particularly in preserving object edge continuity and fine structural details under extreme illumination degradation. In this paper, we propose a novel model, DRWKV (Detailed Receptance Weighted Key Value), which integrates our proposed Global Edge Retinex (GER) theory, enabling effective decoupling of illumination and edge structures for enhanced edge fidelity. Secondly, we introduce Evolving WKV Attention, a spiral-scanning mechanism tha…

HybridTM: Combining Transformer and Mamba for 3D Semantic Segmentation
Xinyu Wang, Jinghua Hou, Zhe Liu, Yingying Zhu
https://arxiv.org/abs/2507.18575 https://

HybridTM: Combining Transformer and Mamba for 3D Semantic Segmentation
Transformer-based methods have demonstrated remarkable capabilities in 3D semantic segmentation through their powerful attention mechanisms, but the quadratic complexity limits their modeling of long-range dependencies in large-scale point clouds. While recent Mamba-based approaches offer efficient processing with linear complexity, they struggle with feature representation when extracting 3D features. However, effectively combining these complementary strengths remains an open challenge in thi…

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[4/4]:
- A novel approach to navigate the taxonomic hierarchy to address the Open-World Scenarios in Medic...
Soumen Sinha, Tanisha Rana, Rahul Roy

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[3/4]:
- Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning
Bob Zhang, Haoran Li, Tao Zhang, Cilin Yan, Jiayin Cai, Yanbin Hao

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[2/4]:
- TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Spl...
Jianchuan Chen, Jingchuan Hu, Gaige Wang, Zhonghua Jiang, Tiansong Zhou, Zhiwen Chen, Chengfei Lv

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[1/4]:
- Context Diffusion: In-Context Aware Image Generation
Najdenkoska, Sinha, Dubey, Mahajan, Ramanathan, Radenovic

Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention
Yiwen Chen, Zhihao Li, Yikai Wang, Hu Zhang, Qin Li, Chi Zhang, Guosheng Lin
https://arxiv.org/abs/2507.17745

Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention
Recent advances in sparse voxel representations have significantly improved the quality of 3D content generation, enabling high-resolution modeling with fine-grained geometry. However, existing frameworks suffer from severe computational inefficiencies due to the quadratic complexity of attention mechanisms in their two-stage diffusion pipelines. In this work, we propose Ultra3D, an efficient 3D generation framework that significantly accelerates sparse voxel modeling without compromising quali…

Yume: An Interactive World Generation Model
Xiaofeng Mao, Shaoheng Lin, Zhen Li, Chuanhao Li, Wenshuo Peng, Tong He, Jiangmiao Pang, Mingmin Chi, Yu Qiao, Kaipeng Zhang
https://arxiv.org/abs/2507.17744

Yume: An Interactive World Generation Model
Yume aims to use images, text, or videos to create an interactive, realistic, and dynamic world, which allows exploration and control using peripheral devices or neural signals. In this report, we present a preview version of \method, which creates a dynamic world from an input image and allows exploration of the world using keyboard actions. To achieve this high-fidelity and interactive video world generation, we introduce a well-designed framework, which consists of four main components, incl…

A Comprehensive Evaluation Framework for the Study of the Effects of Facial Filters on Face Recognition Accuracy
Kagan Ozturk, Louisa Conwill, Jacob Gutierrez, Kevin Bowyer, Walter J. Scheirer
https://arxiv.org/abs/2507.17729

A Comprehensive Evaluation Framework for the Study of the Effects of Facial Filters on Face Recognition Accuracy
Facial filters are now commonplace for social media users around the world. Previous work has demonstrated that facial filters can negatively impact automated face recognition performance. However, these studies focus on small numbers of hand-picked filters in particular styles. In order to more effectively incorporate the wide ranges of filters present on various social media applications, we introduce a framework that allows for larger-scale study of the impact of facial filters on automated …