Tootfinder

🧾 Multi-Modal Vision vs. Text-Based Parsing: Benchmarking LLM Strategies for Invoice Processing
#software

Multi-Modal Vision vs. Text-Based Parsing: Benchmarking LLM Strategies for Invoice Processing
This paper benchmarks eight multi-modal large language models from three families (GPT-5, Gemini 2.5, and open-source Gemma 3) on three diverse openly available invoice document datasets using zero-shot prompting. We compare two processing strategies: direct image processing using multi-modal capabilities and a structured parsing approach converting documents to markdown first. Results show native image processing generally outperforms structured approaches, with performance varying across mode…

Cowboys reveal the why in Jake Ferguson no-show vs. Broncos https://www.sportingnews.com/us/nfl/dallas-cowboys/news/cowboys-reveal-why-jake-ferguson-no-show-vs-broncos/0bc110fc67da02d4a8d3b7c9

Raiders Meet This Make-or-Break Moment in Week 9 vs. Jaguars https://www.si.com/nfl/raiders/onsi/las-vegas-liam-coen-jacksonville-jaguars-pete-carroll

Applications of the Vendi score in genomic epidemiology
Bjarke Frost Nielsen, Amey P. Pasarkar, Qiqi Yang, Bryan T. Grenfell, Adji Bousso Dieng
https://arxiv.org/abs/2509.22520 …

Applications of the Vendi score in genomic epidemiology
The Vendi score (VS), a diversity metric recently conceived in the context of machine learning, with applications in a wide range of fields, has a few distinct advantages over the metrics commonly used in ecology. It is classification-independent, incorporates abundance information, and has a tunable sensitivity to rare/abundant types. Using rich COVID-19 sequence data as a paradigm, we develop methods for applying the VS to time-resolved sequence data. We show how the VS allows for characteriz…

Williams' career day shows Bears proof of concept https://www.espn.com/nfl/story/_/id/46342803/caleb-williams-career-day-bears-show-proof-concept

from my link log —
Flat design vs rich (“skeuomorphic”) design: a bibliography.
https://www.flatisbad.com/
saved 2025-07-30 https://dot…

Willkommen zur einzigen KI-Show mit einer besserwisserischen Kater (Pünktchen) und einem schnarchenden Mops (Lars). Wir streiten uns über:
✅ KI-Suche vs. Menschheit („Lars googelt noch wie 2005!“)
✅ KI-Butler, die bald deinen Urlaub buchen (OpenAI plant einen App Store dafür!)
✅ TikTok-Überflutung mit KI-Clips (wir machen mit – manchmal schlecht)
entscheidet. Und nein, wir lassen uns nicht von Algorithmen die Zahnbürste aussuchen.

Docs: Oracle generated ~$900M from its Nvidia cloud server rental business, with a $125M gross profit, or a 14% margin, vs. its ~70% overall gross profit margin (The Information)
https://www.theinformation.com/articles/internal-oracl…

Internal Oracle Data Show Financial Challenge of Renting Out Nvidia Chips
Oracle became the best-performing megacap stock of 2025 after its executives said last month that the once-sleepy database firm will generate an astonishing $381 billion in revenue from renting out specialized cloud servers to OpenAI and other artificial intelligence developers over the next ...

Will Patriots QB Drake Maye show improvement in Week 2 vs. Dolphins? https://www.espn.com/nfl/story/_/id/46237447/will-new-england-patriots-qb-drake-maye-show-improvement-week-2-vs-dolphins

#WWE #SmackDown is live from the 3Arena in Dublin, #Ireland next week.
These are currently advertised for the show (more will be added throughout the week)
* John Cena returns
* Aleister Black vs R-…

4 things Mazi Smith must show vs Ravens to prove he's capable of playing in Eberflus' D https://cowboyswire.usatoday.com/story/sports/nfl/cowboys/2025/08/15/4-things-to-show-cowboys-mazi-smit…

Scouting Indiana QB Fernando Mendoza: What did 2026 NFL Draft prospect show vs. Illinois? https://www.nytimes.com/athletic/6652054/2025/09/23/nfl-draft-2026-fernando-mendoza-film-illinois/

The Colts Show Mailbag: Bouncing back vs. Raiders, Michael Pittman Jr.'s touchdown total, what's next for Adonai Mitchell https://www.colts.com/news/the-colts-show-mailbag-bouncing-back-vs-raide…

Quasi-conformal VS quasi-isometric equivalence in spaces with controlled growth
Katrin F\"Assler, Enrico Le Donne, Sebastiano Nicolussi Golo, Alessandro Ottazzi, Pierre Pansu
https://arxiv.org/abs/2510.12161

Quasi-conformal VS quasi-isometric equivalence in spaces with controlled growth
We study conditions under which quasi-conformal homeomorphisms are quasi-isometries. We show that if two nilpotent geodesic Lie groups are quasi-conformally homeomorphic, then they are quasi-isometrically equivalent. We also give more general results beyond the nilpotent case. In particular, we show that quasi-conformal homeomorphisms between geodesic Lie groups are quasi-isometries whenever the spaces have strict parabolic or hyperbolic conformal type. As a consequence, quasi-conformal homeomo…

Hues and Cues: Human vs. CLIP
Nuria Alabau-Bosque, Jorge Vila-Tom\'as, Paula Daud\'en-Oliver, Pablo Hern\'andez-C\'amara, Jose Manuel Ja\'en-Lorites, Valero Laparra, Jes\'us Malo
https://arxiv.org/abs/2509.02305

Hues and Cues: Human vs. CLIP
Playing games is inherently human, and a lot of games are created to challenge different human characteristics. However, these tasks are often left out when evaluating the human-like nature of artificial models. The objective of this work is proposing a new approach to evaluate artificial models via board games. To this effect, we test the color perception and color naming capabilities of CLIP by playing the board game Hues & Cues and assess its alignment with humans. Our experiments show that …

Little Caesars Pizza!Pizza! Pregame Show | Week 1 vs. Cowboys https://www.philadelphiaeagles.com/video/little-caesars-pizza-pizza-pregame-show-week-1-vs-cowboys-2025

Double Descent as a Lens for Sample Efficiency in Autoregressive vs. Discrete Diffusion Models
Ahmad Fraij, Sam Dauncey
https://arxiv.org/abs/2509.24974 https://

Double Descent as a Lens for Sample Efficiency in Autoregressive vs. Discrete Diffusion Models
Data scarcity drives the need for more sample-efficient large language models. In this work, we use the double descent phenomenon to holistically compare the sample efficiency of discrete diffusion and autoregressive models. We show that discrete diffusion models require larger capacity and more training epochs to escape their underparameterized regime and reach the interpolation threshold. In the strongly overparameterized regime, both models exhibit similar behavior, with neither exhibiting a…

The anisotropic Heisenberg model close to the Ising limit: triangular lattice vs. effective models
Martin Ulaga, Jure Kokalj, Takami Tohyama, Peter Prelov\v{s}ek
https://arxiv.org/abs/2510.12667

The anisotropic Heisenberg model close to the Ising limit: triangular lattice vs. effective models
Stimulated by recent experiments on materials representing the realization of the anisotropic Heisenberg spin-$1/2$ model on the triangular lattice, we explore further properties of such a model in the easy-axis regime $α= J_\perp/J_z < 1$, as well as effective models that also capture such physics. We show that anisotropic Heisenberg models on the honeycomb lattice and even on the square lattice reveal similarities to the full triangular lattice in the magnetization curve as well as in the tr…

Long Polar vs. LDPC Codes under Complexity-Constrained Decoding
Felix Krieg, Marvin R\"ubenacke, Andreas Zunker, Stephan ten Brink
https://arxiv.org/abs/2508.05485 https://…

Long Polar vs. LDPC Codes under Complexity-Constrained Decoding
The prevailing opinion in industry and academia is that polar codes are competitive for short code lengths, but can no longer keep up with low-density parity-check (LDPC) codes as block length increases. This view is typically based on the assumption that LDPC codes can be decoded with a large number of belief propagation (BP) iterations. However, in practice, the number of iterations may be rather limited due to latency and complexity constraints. In this paper, we show that for a similar numb…

The Hidden DNA of LLM-Generated JavaScript: Structural Patterns Enable High-Accuracy Authorship Attribution
Norbert Tihanyi, Bilel Cherif, Richard A. Dubniczky, Mohamed Amine Ferrag, Tam\'as Bisztray
https://arxiv.org/abs/2510.10493

The Hidden DNA of LLM-Generated JavaScript: Structural Patterns Enable High-Accuracy Authorship Attribution
In this paper, we present the first large-scale study exploring whether JavaScript code generated by Large Language Models (LLMs) can reveal which model produced it, enabling reliable authorship attribution and model fingerprinting. With the rapid rise of AI-generated code, attribution is playing a critical role in detecting vulnerabilities, flagging malicious content, and ensuring accountability. While AI-vs-human detection usually treats AI as a single category we show that individual LLMs le…

Data from 300K pull requests shows OpenAI is catching up to Anthropic in AI coding: Codex has a 74.3% success rate vs. Claude Code's 73.7% in code approvals (Stephanie Palazzolo/The Information)
https://www.theinformation.com/articles/openai-catching-anthropi…

OpenAI Is Catching Up To Anthropic in AI Coding
OpenAI’s effort to catch up to Anthropic in code-generating artificial intelligence seems to be working.New data show OpenAI’s Codex coding assistant has pulled ahead of Anthropic’s Claude Code assistant in certain coding capabilities. Codex usage among developers is also catching up to Claude ...

Holography vs. Scale Separation
Alek Bedroya, Paul J. Steinhardt
https://arxiv.org/abs/2509.25313 https://arxiv.org/pdf/2509.25313

Holography vs. Scale Separation
In this work, we point out a contradiction between holography and scale-separated AdS (i.e. parametrically large mass gap) in string theory, making the standard assumption that the holographic CFT describes the IR degrees of freedom on a brane that decouple from gravity. We show that the CFT can only decouple from gravity if the scalar potential in the dual AdS satisfies a certain criterion. Namely, there must exist a scalar field trajectory that follows the gradient of the scalar potential to …

Will Annotators Disagree? Identifying Subjectivity in Value-Laden Arguments
Amir Homayounirad, Enrico Liscio, Tong Wang, Catholijn M. Jonker, Luciano C. Siebert
https://arxiv.org/abs/2509.06704

Will Annotators Disagree? Identifying Subjectivity in Value-Laden Arguments
Aggregating multiple annotations into a single ground truth label may hide valuable insights into annotator disagreement, particularly in tasks where subjectivity plays a crucial role. In this work, we explore methods for identifying subjectivity in recognizing the human values that motivate arguments. We evaluate two main approaches: inferring subjectivity through value prediction vs. directly identifying subjectivity. Our experiments show that direct subjectivity identification significantly …

RobQFL: Robust Quantum Federated Learning in Adversarial Environment
Walid El Maouaki, Nouhaila Innan, Alberto Marchisio, Taoufik Said, Muhammad Shafique, Mohamed Bennai
https://arxiv.org/abs/2509.04914

RobQFL: Robust Quantum Federated Learning in Adversarial Environment
Quantum Federated Learning (QFL) merges privacy-preserving federation with quantum computing gains, yet its resilience to adversarial noise is unknown. We first show that QFL is as fragile as centralized quantum learning. We propose Robust Quantum Federated Learning (RobQFL), embedding adversarial training directly into the federated loop. RobQFL exposes tunable axes: client coverage $γ$ (0-100\%), perturbation scheduling (fixed-$\varepsilon$ vs $\varepsilon$-mixes), and optimization (fine-tun…

Top Draft Prospects to Watch in Alabama vs. Wisconsin https://insidethestar.com/top-draft-prospects-to-watch-in-alabama-vs-wisconsin

Characterizing the host galaxies and delay times of Ca-rich gap transients vs 91bg-like SNe and normal Type Ia SNe
Peter Scherbak, Abigail Polin, Mansi Kasliwal, Kishaley De, Peter Behroozi, Dave Cook, Wynn Jacobson-Gal\'an
https://arxiv.org/abs/2509.25577

Characterizing the host galaxies and delay times of Ca-rich gap transients vs 91bg-like SNe and normal Type Ia SNe
Calcium-rich gap transients are a faint, fast-evolving class of supernovae that show strong nebular Ca emission lines. Their progenitor systems are uncertain, but they are often associated with old and quiescent host galaxies. In this work, we compare the properties of the hosts of hydrogen-poor Ca-rich gap transients to the hosts of 3 other classes of supernova (SNe): normal Type Ia, 91bg-like, and Type II. We use data from the Zwicky Transient Facility (ZTF) Census of the Local Universe (CLU)…

I won't be surprised if the US President live tweets this show.
#VictoriasSecret #fashion #USA #inevitable #VS_Show

Patrick Graham: 'It's all about how we prepare and show up Sunday' https://www.raiders.com/video/patrick-graham-raiders-week-1-vs-patriots-presser-nfl-09042025

Brandon Graham previews Eagles Week 1 matchup vs. Cowboys https://www.nfl.com/videos/brandon-graham-previews-eagles-week-1-matchup-vs-cowboys-gmfb

The Critical Adjustment Raiders Must Make vs. Chargers https://www.si.com/nfl/raiders/las-vegas-ashton-jeanty-chip-kelly-los-angeles-chargers-pete-carroll

Ferromagnetism vs. Antiferromagnetism in Narrow-Band Systems: Competition Between Quantum Geometry and Band Dispersion
Haoyu Hu, Oskar Vafek, Kristjan Haule, B. Andrei Bernevig
https://arxiv.org/abs/2509.03575

Ferromagnetism vs. Antiferromagnetism in Narrow-Band Systems: Competition Between Quantum Geometry and Band Dispersion
Magnetism in narrow-band systems arises from the interplay between electronic correlations, quantum geometry, and band dispersion. In particular, both ferro and anti-ferro magnets are known to occur as ground states of (different) models featuring narrow bands. This poses the question of which is favored and under what conditions. In this work, we present a unified theoretical framework to investigate spin physics within narrow bands. By deriving an effective spin model, we show that the non-at…

Giants Huddle https://www.giants.com/video/giants-huddle-dave-helman