Tootfinder

cora: CORA citations (1998)
Citations among papers indexed by CORA, from 1998, an early computer science research paper search engine. If a paper i cites a paper j also in this data set, then a directed edge connects i to j. (Papers not in the data set are excluded.) Self-loops may be present. The dates of these snapshots are uncertain.
This network has 23166 nodes and 91500 edges.
Tags: Informational, Citation, Unweighted

Measuring #EnergyConsumption in #ProgrammingLanguages for #AI Applications

Gotta admit… there's nothing quite like the feel of holding a paper copy of your book in your hands. (First time since 2007, haven't done new printing version since.) And color print-on-demand has gotten remarkably affordable, so price unchanged (just drop royalty a bit).

Remaining Time Prediction in Outbound Warehouse Processes: A Case Study (Short Paper)
Erik Penther, Michael Grohs, Jana-Rebecca Rehse
https://arxiv.org/abs/2509.18986 https://…

🩻 Pictures from Paper Reflections and a Single Pixel
#makers

Pictures From Paper Reflections And A Single Pixel
Taking a picture with a single photoresistor is a brain-breaking idea. But go deeper and imagine taking that same picture with the same photoresistor, but without even facing the object. [Jon Bumst…

"The Scale of Russian Sabotage Operations Against Europe’s Critical Infrastructure" by IISS.
https://www.iiss.org/research-paper/2025/08/the-scale-of-russian--sabotage-operations--against-europes-critic…

The Scale of Russian Sabotage Operations Against Europe’s Critical Infrastructure
This IISS paper assesses Russia’s unconventional war on Europe, focusing on sabotage of critical infrastructure, from military sites and energy grids to communications and undersea cables, testing the resilience of European governments and societies and challenging NATO/EU deterrence.

Bounded ribbonlength for knot families and multi-twist M\"obius bands
Elizabeth Denne, Timi Patterson
https://arxiv.org/abs/2509.18370 https://arxiv.o…

Bounded ribbonlength for knot families and multi-twist Möbius bands
Take a thin, rectangular strip of paper, add in an odd number of half-twists, then join the ends together. This gives a multi-twist paper Möbius band. We prove that any multi-twist paper Möbius band can be constructed so the aspect ratio of the rectangle is $3\sqrt{3}+ε$ for any $ε>0$. We could also take the thin, rectangular strip of paper and tie a knot in it, then join the ends and fold flat in the plane. This creates a folded ribbon knot. We apply the techniques used to prove the multi-…

Learning to Write on Dirty Paper
Ezgi Ozyilkan, O\u{g}uzhan Kubilay \"Ulger, Elza Erkip
https://arxiv.org/abs/2507.17427 https://arxiv.org/pdf/2507.17…

Learning to Write on Dirty Paper
Dirty paper coding (DPC) is a classical problem in information theory that considers communication in the presence of channel state known only at the transmitter. While the theoretical impact of DPC has been substantial, practical realizations of DPC, such as Tomlinson-Harashima precoding (THP) or lattice-based schemes, often rely on specific modeling assumptions about the input, state and channel. In this work, we explore whether modern learning-based approaches can offer a complementary path …

Navigation through Non-Compact Symmetric Spaces: a mathematical perspective on Cartan Neural Networks
Pietro Giuseppe Fr\'e, Federico Milanesio, Guido Sanguinetti, Matteo Santoro
https://arxiv.org/abs/2507.16871

Navigation through Non-Compact Symmetric Spaces: a mathematical perspective on Cartan Neural Networks
Recent work has identified non-compact symmetric spaces U/H as a promising class of homogeneous manifolds to develop a geometrically consistent theory of neural networks. An initial implementation of these concepts has been presented in a twin paper under the moniker of Cartan Neural Networks, showing both the feasibility and the performance of these geometric concepts in a machine learning context. The current paper expands on the mathematical structures underpinning Cartan Neural Networks, de…

Two-flow Feedback Multi-scale Progressive Generative Adversarial Network
Sun Weikai, Song Shijie, Chi Wenjie
https://arxiv.org/abs/2508.16089 https://arxiv…

Two-flow Feedback Multi-scale Progressive Generative Adversarial Network
Although diffusion model has made good progress in the field of image generation, GAN\cite{huang2023adaptive} still has a large development space due to its unique advantages, such as WGAN\cite{liu2021comparing}, SSGAN\cite{guibas2021adaptive} \cite{zhang2022vsa} \cite{zhou2024adapt} and so on. In this paper, we propose a novel two-flow feedback multi-scale progressive generative adversarial network (MSPG-SEN) for GAN models. This paper has four contributions: 1) : We propose a two-flow feedbac…

Connecting Quantum Computing with Classical Stochastic Simulation
Jose Blanchet, Mark S. Squillante, Mario Szegedy, Guanyang Wang
https://arxiv.org/abs/2509.18614 https://

citeseer: CiteSeer citations (2014)
Citations among papers indexed by the CiteSeer digital library. If a paper i cites a paper j also in this data set, then a directed edge connects i to j. (Papers not in the data set are excluded.) Self-loops may be present.
This network has 384413 nodes and 1751463 edges.
Tags: Informational, Citation, Unweighted

Rigby keeps trying to invent money, but he doesn't have a coherent model of interest rate policy in response to tariff shocks, so I'm not convinced.

Genuinely interested in any thoughts as to where to send it next. The subject is Antarctica and influences on observed ice sheet change
#academicchatter

Weaving the Future: Generative AI and the Reimagining of Fashion Design
Pierre-Marie Chauvin (IFM, AHP-PReST), Ang\`ele Merlin (IFM, AHP-PReST), Xavier Fresquet (IFM, AHP-PReST), Hugo Caselles-Dupr\'e (IFM, AHP-PReST), Benjamin Simmenauer (IFM, AHP-PReST), Mathieu de Fayet
https://arxiv.org/abs/2507.17758…

Weaving the Future: Generative AI and the Reimagining of Fashion Design
This paper explores the integration of generative AI into the fashion design process. Drawing on insights from the January 2025 seminar ``Tisser le futur,'' it investigates how AI reshapes creative workflows, from ideation to prototyping, while interrogating the ethical, aesthetic, and labor implications. The paper highlights co-creative dynamics between humans and machines, the potential for aesthetic innovation, and the environmental and cultural challenges of algorithmic design.

Sources: all but one person on WaPo's obit desk has taken the paper's buyout offer; the lone holdout is too junior to qualify for the buyout deal (Michael Schaffer/Politico)
https://www.politico.com/news/magazine/2025/07/24/washin…

A Stateless Transparent Voting Machine
Juan E. Gilbert, Jean D. Louis
https://arxiv.org/abs/2509.19257 https://arxiv.org/pdf/2509.19257

I'm reading the latest issue of Classic Pop Magazine (yes, an actual magazine made of actual paper and with actual pages & stuff) and in the contents I see this image.
Well, you know me. There is literally only one thing I can say right now:
KEYTAR KLAXON!
#keytar #totp #80sMusic

Motivic classes of fixed-generators Hilbert schemes of unibranch curve singularities and Igusa zeta functions
Ilaria Rossinelli
https://arxiv.org/abs/2507.17642 https://

Motivic classes of fixed-generators Hilbert schemes of unibranch curve singularities and Igusa zeta functions
This paper delves into the study of Hilbert schemes of unibranch plane curves whose points have a fixed number of minimal generators. Building on the work of Oblomkov, Rasmussen and Shende we provide a formula for their motivic classes and investigate the relationship with principal Hilbert schemes of the same given unibranch curve. In addition, the paper specializes this study to the case of $(p,q)$-curves, where we obtain more structured results for the motivic classes of fixed-generators Hil…

A zoo of coisotropic embeddings
Rub\'en Izquierdo-L\'opez, Manuel de Le\'on, Luca Schiavone, Pablo Soto
https://arxiv.org/abs/2509.19039 https://

Variable Neighborhood Descent Methods for Large-scale Single-assignment Multi-level Facility Location Problem
Haibo Wang, Bahram Alidaee
https://arxiv.org/abs/2508.15954 https:/…

Variable Neighborhood Descent Methods for Large-scale Single-assignment Multi-level Facility Location Problem
This paper addresses the single--assignment uncapacitated multi-level facility location (MFL) problem, which has numerous applications, including tactical and strategic supply chain management. We consider four-level and five-level facilities (4-LFL and 5-LFL). Although the MFL has been addressed in the literature in various settings, solutions to large-scale, realistic problems are still lacking. This paper considers several variants of the variable neighborhood descent (VND) method, including…

I can't believe that retraction took 15 years – and that the authors still haven't accepted their L. That study was extremely oversold from the get-go, by the author's not being self-critical enough and wanting to believe their own extraordinary claims.
«Controversial ‘arsenic life’ paper retracted after 15 years — but authors fight back»
https://www.nature.com/articles/d41586-025-02325-z

Oh, Hacker News…FFS. Something got lost in the sad game of Popular Press Telephone between the actual study and this snotty post.
The actual study is here:
https://www.acpjournals.org/doi/10.7326/ANNALS-25-00577
It notes that:
1. Most research on air cleaning devices used air samples, not human health outcomes. (It’s far more difficult to conduct a study on the latter.)
2. Not many people are studying whether e.g. UV air purifiers create toxic byproducts.
3. The available studies sure do measure a whole bunch of different things, and wouldn’t it be nice if they had more consistent standards.
Their main recommendation is basically “scientists should do more studies where they measure infection in humans, not just germs in the air.” And that’s fair: It’s reasonable to •guess• that fewer germs in the air means less sickness, but it’s nice to actually •test• that!
However, AFAICT, nothing in the study warrants the “tinfoil hats” remark. Unlike a tinfoil hats, we have plenty good reason to think at least some of these devices actually work; the paper just wants the research to go deeper. It’s one giant “further study needed.” https://mastodon.social/@ngate/115091700894746240

Parameter Height bounds for the Zilber Pink conjecture for PEL types III and IV
Bijay Raj Bhatta
https://arxiv.org/abs/2507.16827 https://

Parameter Height bounds for the Zilber Pink conjecture for PEL types III and IV
We prove the Zilber-Pink conjecture to the intersection of an irreducible Hodge generic algebraic subvariety $ V \subset \mathcal{A}_g$ with special subvarieties of all simple PEL types other than $\mathbb{Z}$, under the assumption of the Large Galois Orbits conjecture. In particular, we establish parameter height bounds for the arithmetic ingredients of the Pila-Zannier strategy in the case of Albert types III and IV. This paper is a sequel to Daw and Orr's paper "Lattices with skew-Hermitian …

The last step in a long process on “arsenic life”: #arsenic Instead of phosphorus” from 2010 has now been killed (over the objections of the authors) by the journal that had published it: https://www.science.org/doi/10.1126/science.adu5488 (and see https://skyweek.wordpress.com/2010/12/02/bakterien-mit-arsen-zwangsernahrt-macht-das-leben-im-all-wahrscheinlicher/ and https://skyweek.wordpress.com/2010/12/05/zweifel-an-den-arsen-bakterien-mehren-sich/ for my reporting back then about the announcement and the swift and hard criticism).

There is a roughly 1991 Dodge Dakota parked in the neighborhood that I am yet to see move at all in the fifteen years I've lived here. I know that it's probably held together with paper mache, but that is one sexy looking truck.

Let's Use ChatGPT To Write Our Paper! Benchmarking LLMs To Write the Introduction of a Research Paper
Krishna Garg, Firoz Shaikh, Sambaran Bandyopadhyay, Cornelia Caragea
https://arxiv.org/abs/2508.14273

Let's Use ChatGPT To Write Our Paper! Benchmarking LLMs To Write the Introduction of a Research Paper
As researchers increasingly adopt LLMs as writing assistants, generating high-quality research paper introductions remains both challenging and essential. We introduce Scientific Introduction Generation (SciIG), a task that evaluates LLMs' ability to produce coherent introductions from titles, abstracts, and related works. Curating new datasets from NAACL 2025 and ICLR 2025 papers, we assess five state-of-the-art models, including both open-source (DeepSeek-v3, Gemma-3-12B, LLaMA 4-Maverick, Mi…

dblp_cite: DBLP citations (2014)
Citations among papers contained in the DBLP computer science bibliography. If a paper i cites a paper j also in this data set, then a directed edge connects i to j. (Papers not in the data set are excluded.) Self-loops may be present. This snapshot from May 2014.
This network has 12590 nodes and 49759 edges.
Tags: Informational, Citation, Unweighted

29 Jul 2025 webinar: That Doesn't Need to Be a PDF
Hear how agencies can design an HTML-first approach for electronic content using “@media print” for a superior printing experience (because yes, some people do use paper) during a panel discussion with experts from FDIC, USAB, and GSA when a PDF document is the proper solution.
h…

Position Paper: Integrating Explainability and Uncertainty Estimation in Medical AI
Xiuyi Fan
https://arxiv.org/abs/2509.18132 https://arxiv.org/pdf/2509.1…

Position Paper: Integrating Explainability and Uncertainty Estimation in Medical AI
Uncertainty is a fundamental challenge in medical practice, but current medical AI systems fail to explicitly quantify or communicate uncertainty in a way that aligns with clinical reasoning. Existing XAI works focus on interpreting model predictions but do not capture the confidence or reliability of these predictions. Conversely, uncertainty estimation (UE) techniques provide confidence measures but lack intuitive explanations. The disconnect between these two areas limits AI adoption in medi…

I got completely sidetracked by client work for the past week but I did finish my last batch of hand made recycled paper.
#diy #maker #paper

#Blakes7 Series B, Episode 07 - Killer
TYNUS: Nervous?
AVON: Just careful.
TYNUS: We've just received this odd message. I think it's from your friends on the Liberator.
https://blake.torpidity.net/m/207/128

Consistency of Selection Strategies for Fraud Detection
Christos Revelas, Otilia Boldea, Bas J. M. Werker
https://arxiv.org/abs/2509.18739 https://arxiv.or…

Partial sharing and cross sharing of entire function with its derivative
Sujoy Majumder, Nabadwip Sarkar, Debabrata Pramanik
https://arxiv.org/abs/2509.20273 https://

Partial sharing and cross sharing of entire function with its derivative
In the paper, we investigate the uniqueness problem of a power of an entire function that share one value partially with it's linear differential polynomial and obtain a result, which improves several previous results in a large scale. Also in the paper we include some applications of our main result. Moreover, we solve the question raised by Wang and Liu (Value cross-sharing of meromorphic functions, Comput. Methods Funct. Theory (2023). https://doi.org/10.1007/s40315-023-00481-9) for a specia…

Vertex-minimal hyperbolic origami 2-torus
Zhengyu Zou
https://arxiv.org/abs/2509.18668 https://arxiv.org/pdf/2509.18668

Vertex-minimal hyperbolic origami 2-torus
In this paper, we explore geometrizations of triangulated hyperbolic 2-tori $Σ_2$ in the hyperbolic 3-space $\mathbb{H}^3$. We show that for all $n \geq 10$, there exists a triangulation of $Σ_2$ with $n$ vertices and an embedding $S: Σ_2 \hookrightarrow \mathbb{H}^3$ that restricts to isometries on the triangles. We call these embeddings hyperbolic origami 2-tori. Since the lower bound $n \geq 10$ is combinatorially sharp, this paper settles the question of minimum number of vertices requir…

An interview with Alex Kachkine, who shook the art world in June with a research paper in Nature on restoring paintings using AI, as he works to refine the tech (Ephrat Livni/New York Times)
https://www.nytimes.com/2025/08/…

The Hobbyist Restorer Who Rocked the Art World With an A.I. Innovation
Alex Kachkine spends his days working on microchip research — a skill set surprisingly similar to that needed for restoration.

The Proof-Theoretic Origin of Double Negation Introduction & Elimination
Khashayar Irani
https://arxiv.org/abs/2509.17623 https://arxiv.org/pdf/2509.17…

The Proof-Theoretic Origin of Double Negation Introduction & Elimination
This paper investigates the proof-theoretic foundations of double negation introduction (DNI) and double negation elimination (DNE) in classical logic. By examining both sequent calculus and natural deduction, it is shown that these rules originate in reductio ad absurdum. The paper demonstrates that both rules possess harmony, ensuring balance between introduction and elimination, and normalisation, which guarantees that derivations reduce to canonical form without detours. These features reve…

How to Beat Nakamoto in the Race
Shu-Jie Cao, Dongning Guo
https://arxiv.org/abs/2508.16202 https://arxiv.org/pdf/2508.16202…

How to Beat Nakamoto in the Race
This paper studies proof-of-work Nakamoto consensus under bounded network delays, settling two long-standing questions in blockchain security: How can an adversary most effectively attack block safety under a given block confirmation latency? And what is the resulting probability of safety violation? A Markov decision process (MDP) framework is introduced to precise characterize the system state (including the tree and timings of all blocks mined), the adversary's potential actions, and the sta…

G2S-ICP SLAM: Geometry-aware Gaussian Splatting ICP SLAM
Gyuhyeon Pak, Hae Min Cho, Euntai Kim
https://arxiv.org/abs/2507.18344 https://arxiv.org/pdf/2507.…

G2S-ICP SLAM: Geometry-aware Gaussian Splatting ICP SLAM
In this paper, we present a novel geometry-aware RGB-D Gaussian Splatting SLAM system, named G2S-ICP SLAM. The proposed method performs high-fidelity 3D reconstruction and robust camera pose tracking in real-time by representing each scene element using a Gaussian distribution constrained to the local tangent plane. This effectively models the local surface as a 2D Gaussian disk aligned with the underlying geometry, leading to more consistent depth interpretation across multiple viewpoints comp…

35 years ago #OTD the #Hubble Space Telescope with its mishappen optics took this image of #Saturn - which could be sharpened a lot mathematically (as the paper https://articles.adsabs.harvard.edu/full/1991ApJ...369L..51W later described) since the object was so bright. It's being said that this one image was used by NASA to convince key U.S. politicians that the broken satellite still had some life in it ... and so it was, for 35 years now!

Non-Asymptotic Achievable Rate-Distortion Region for Indirect Wyner-Ziv Source Coding
Jiahui Wei, Philippe Mary, Elsa Dupraz
https://arxiv.org/abs/2507.17432 https://

Non-Asymptotic Achievable Rate-Distortion Region for Indirect Wyner-Ziv Source Coding
In the Wyner-Ziv source coding problem, a source $X$ has to be encoded while the decoder has access to side information $Y$. This paper investigates the indirect setup, in which a latent source $S$, unobserved by both the encoder and the decoder, must also be reconstructed at the decoder. This scenario is increasingly relevant in the context of goal-oriented communications, where $S$ can represent semantic information obtained from $X$. This paper derives the indirect Wyner-Ziv rate-distortion …

Sparse optimal control for infinite-dimensional linear systems with applications to graphon control
Takuya Ikeda, Masaaki Nagahara
https://arxiv.org/abs/2507.18030 https://

Sparse optimal control for infinite-dimensional linear systems with applications to graphon control
Large-scale networked systems typically operate under resource constraints, and it is also difficult to exactly obtain the network structure between nodes. To address these issues, this paper investigates a sparse optimal control for infinite-dimensional linear systems and its application to networked systems where the network structure is represented by a limit function called a graphon that captures the overall connection pattern. The contributions of this paper are twofold: (i) To reduce com…

Failure Modes of Maximum Entropy RLHF
\"Omer Veysel \c{C}a\u{g}atan, Bar{\i}\c{s} Akg\"un
https://arxiv.org/abs/2509.20265 https://arxiv.org/pdf/…

Failure Modes of Maximum Entropy RLHF
In this paper, we show that Simple Preference Optimization (SimPO) can be derived as Maximum Entropy Reinforcement Learning with length-normalized temperature, providing a theoretical foundation for this reference-free method. Motivated by SimPO's strong performance in offline preference optimization, we investigate whether Maximum Entropy RL can achieve similar results in online RLHF settings. Our experiments find that Maximum Entropy RL consistently exhibits overoptimization and unstable KL d…

Attention Mechanism in Randomized Time Warping
Yutaro Hiraoka, Kazuya Okamura, Kota Suto, Kazuhiro Fukui
https://arxiv.org/abs/2508.16366 https://arxiv.org…

Attention Mechanism in Randomized Time Warping
This paper reveals that we can interpret the fundamental function of Randomized Time Warping (RTW) as a type of self-attention mechanism, a core technology of Transformers in motion recognition. The self-attention is a mechanism that enables models to identify and weigh the importance of different parts of an input sequential pattern. On the other hand, RTW is a general extension of Dynamic Time Warping (DTW), a technique commonly used for matching and comparing sequential patterns. In essence,…

Thinking Augmented Pre-training
Liang Wang, Nan Yang, Shaohan Huang, Li Dong, Furu Wei
https://arxiv.org/abs/2509.20186 https://arxiv.org/pdf/2509.20186

Thinking Augmented Pre-training
This paper introduces a simple and scalable approach to improve the data efficiency of large language model (LLM) training by augmenting existing text data with thinking trajectories. The compute for pre-training LLMs has been growing at an unprecedented rate, while the availability of high-quality data remains limited. Consequently, maximizing the utility of available data constitutes a significant research challenge. A primary impediment is that certain high-quality tokens are difficult to le…

Neighbor Embeddings Using Unbalanced Optimal Transport Metrics
Muhammad Rana, Keaton Hamm
https://arxiv.org/abs/2509.19226 https://arxiv.org/pdf/2509.19226…

If you were really lucky they'd end up publishing a paper on the phenomenon. So, are the stories about gangs, violent crime and explosions in Europe just a coincidence in my feed? Or is something really going on? There's no question that the Netherlands underworld has turned very nasty.for instance.

On Deligne's conjecture for Hecke characters
Yubo Jin, Dongwen Liu, Binyong Sun
https://arxiv.org/abs/2509.18921 https://arxiv.org/pdf/2509.18921

Ergodic Properties of Quantum Markov Semigroups
Nicolas Mousset, Nina H. Amini
https://arxiv.org/abs/2509.20133 https://arxiv.org/pdf/2509.20133

Ergodic Properties of Quantum Markov Semigroups
In this paper, we study the ergodic theorem for infinite-dimensional quantum Markov semigroups, originally introduced by Frigerio and Verri in 1982, and its latest version developed by Carbone and Girotti in 2021. We provide a sufficient condition that ensures exponential convergence towards the positive recurrent subspace, a well-known result for irreducible quantum Markov semigroups in finite-dimensional Hilbert spaces. Several illustrative examples are presented to demonstrate the applicatio…

Robot-mediated physical Human-Human Interaction in Neurorehabilitation: a position paper
Lorenzo Vianello, Matthew Short, Julia Manczurowsky, Emek Bar{\i}\c{s} K\"u\c{c}\"uktabak, Francesco Di Tommaso, Alessia Noccaro, Laura Bandini, Shoshana Clark, Alaina Fiorenza, Francesca Lunardini, Alberto Canton, Marta Gandolla, Alessandra L. G. Pedrocchi, Emilia Ambrosini, Manuel Murie-Fernandez, Carmen B. Roman, Jesus Tornero, Natacha Leon, Andrew Sawers, Jim Patton, Domenico Formica,…

cora: CORA citations (1998)
Citations among papers indexed by CORA, from 1998, an early computer science research paper search engine. If a paper i cites a paper j also in this data set, then a directed edge connects i to j. (Papers not in the data set are excluded.) Self-loops may be present. The dates of these snapshots are uncertain.
This network has 23166 nodes and 91500 edges.
Tags: Informational, Citation, Unweighted

Georgia State University’s Center for High Angular Resolution Astronomy (CHARA) Array is marking the 20th anniversary of its first scientific results, published in July 2005: #CHARA interferometer enabled the discoveries.

Track-On2: Enhancing Online Point Tracking with Memory
G\"orkay Aydemir, Weidi Xie, Fatma G\"uney
https://arxiv.org/abs/2509.19115 https://arxiv.…

Track-On2: Enhancing Online Point Tracking with Memory
In this paper, we consider the problem of long-term point tracking, which requires consistent identification of points across video frames under significant appearance changes, motion, and occlusion. We target the online setting, i.e. tracking points frame-by-frame, making it suitable for real-time and streaming applications. We extend our prior model Track-On into Track-On2, a simple and efficient transformer-based model for online long-term tracking. Track-On2 improves both performance and ef…

cora: CORA citations (1998)
Citations among papers indexed by CORA, from 1998, an early computer science research paper search engine. If a paper i cites a paper j also in this data set, then a directed edge connects i to j. (Papers not in the data set are excluded.) Self-loops may be present. The dates of these snapshots are uncertain.
This network has 23166 nodes and 91500 edges.
Tags: Informational, Citation, Unweighted

Bounded PCTL Model Checking of Large Language Model Outputs
Dennis Gross, Helge Spieker, Arnaud Gotlieb
https://arxiv.org/abs/2509.18836 https://arxiv.org/…

Bounded PCTL Model Checking of Large Language Model Outputs
In this paper, we introduce LLMCHECKER, a model-checking-based verification method to verify the probabilistic computation tree logic (PCTL) properties of an LLM text generation process. We empirically show that only a limited number of tokens are typically chosen during text generation, which are not always the same. This insight drives the creation of $α$-$k$-bounded text generation, narrowing the focus to the $α$ maximal cumulative probability on the top-$k$ tokens at every step of the tex…

Diffusion-Modeled Reinforcement Learning for Carbon and Risk-Aware Microgrid Optimization
Yunyi Zhao, Wei Zhang, Cheng Xiang, Hongyang Du, Dusit Niyato, Shuhua Gao
https://arxiv.org/abs/2507.16867

Diffusion-Modeled Reinforcement Learning for Carbon and Risk-Aware Microgrid Optimization
This paper introduces DiffCarl, a diffusion-modeled carbon- and risk-aware reinforcement learning algorithm for intelligent operation of multi-microgrid systems. With the growing integration of renewables and increasing system complexity, microgrid communities face significant challenges in real-time energy scheduling and optimization under uncertainty. DiffCarl integrates a diffusion model into a deep reinforcement learning (DRL) framework to enable adaptive energy scheduling under uncertainty…

WEST: LLM based Speech Toolkit for Speech Understanding, Generation, and Interaction
Binbin Zhang, Chengdong Liang, Shuai Wang, Xuelong Geng, Zhao Guo, Haoyu Li, Hao Yin, Xipeng Yang, Pengshen Zhang, Changwei Ma, Lei Xie
https://arxiv.org/abs/2509.19902

WEST: LLM based Speech Toolkit for Speech Understanding, Generation, and Interaction
In this paper, we present WEST(WE Speech Toolkit), a speech toolkit based on a large language model (LLM) for speech understanding, generation, and interaction. There are three key features of WEST: 1) Fully LLM-based: Standing on the shoulders of giants by reusing mature architectures, ecosystems (e.g., Hugging Face), and methods (e.g., sequence packing) from large models. 2) Full-stack: Supports tasks such as recognition, synthesis, understanding, dialogue, and multimodal capabilities, with e…

The formation and evolution of dust in the colliding-wind binary Apep revealed by JWST: #binary Apep [...] reveal an intricate series of nested concentric dust shells that are abundant in detailed substructure. The striking regularity in these substructures between successive shells suggests an exactly repeating formation mechanism combined with a highly stable outflow that maintains a consistent morphology even after reaching 0.6 pc (assuming a distance of 2.4 kpc) into the interstellar medium." See also https://bsky.app/profile/landru79.bsky.social/post/3luqbnb2jmk25 for other processing experiments with the image, https://arxiv.org/abs/2507.14610 for another paper and https://theconversation.com/swirling-nebula-of-two-dying-stars-revealed-in-spectacular-detail-in-new-webb-telescope-image-258314 for an article!

arxiv_collab: Scientific collaborations in physics (1995-2005)
Collaboration graphs for scientists, extracted from the Los Alamos e-Print arXiv (physics), for 1995-1999 for three categories, and additionally for 1995-2003 and 1995-2005 for one category. For copyright reasons, the MEDLINE (biomedical research) and NCSTRL (computer science) collaboration graphs from this paper are not publicly available.
This network has 8361 nodes and 15751 edges.
Tags: Social, Collaboration…

Robust Small Methane Plume Segmentation in Satellite Imagery
Khai Duc Minh Tran, Hoa Van Nguyen, Aimuni Binti Muhammad Rawi, Hareeshrao Athinarayanarao, Ba-Ngu Vo
https://arxiv.org/abs/2508.16282

Robust Small Methane Plume Segmentation in Satellite Imagery
This paper tackles the challenging problem of detecting methane plumes, a potent greenhouse gas, using Sentinel-2 imagery. This contributes to the mitigation of rapid climate change. We propose a novel deep learning solution based on U-Net with a ResNet34 encoder, integrating dual spectral enhancement techniques (Varon ratio and Sanchez regression) to optimise input features for heightened sensitivity. A key achievement is the ability to detect small plumes down to 400 m2 (i.e., for a single pi…

Rethinking VAE: From Continuous to Discrete Representations Without Probabilistic Assumptions
Songxuan Shi
https://arxiv.org/abs/2507.17255 https://…

Rethinking VAE: From Continuous to Discrete Representations Without Probabilistic Assumptions
This paper explores the generative capabilities of Autoencoders (AEs) and establishes connections between Variational Autoencoders (VAEs) and Vector Quantized-Variational Autoencoders (VQ-VAEs) through a reformulated training framework. We demonstrate that AEs exhibit generative potential via latent space interpolation and perturbation, albeit limited by undefined regions in the encoding space. To address this, we propose a new VAE-like training method that introduces clustering centers to enha…

LLM-Based Agents for Competitive Landscape Mapping in Drug Asset Due Diligence
Alisa Vinogradova (Optic Inc), Vlad Vinogradov (Optic Inc), Dmitrii Radkevich (Optic Inc), Ilya Yasny (Optic Inc), Dmitry Kobyzev (Optic Inc), Ivan Izmailov (Optic Inc), Katsiaryna Yanchanka (Optic Inc), Andrey Doronichev (Optic Inc)
https://arxiv.org/abs/2508.1…

LLM-Based Agents for Competitive Landscape Mapping in Drug Asset Due Diligence
In this paper, we describe and benchmark a competitor-discovery component used within an agentic AI system for fast drug asset due diligence. A competitor-discovery AI agent, given an indication, retrieves all drugs comprising the competitive landscape of that indication and extracts canonical attributes for these drugs. The competitor definition is investor-specific, and data is paywalled/licensed, fragmented across registries, ontology-mismatched by indication, alias-heavy for drug names, mul…

Annif at the GermEval-2025 LLMs4Subjects Task: Traditional XMTC Augmented by Efficient LLMs
Osma Suominen, Juho Inkinen, Mona Lehtinen
https://arxiv.org/abs/2508.15877 https://

Annif at the GermEval-2025 LLMs4Subjects Task: Traditional XMTC Augmented by Efficient LLMs
This paper presents the Annif system in the LLMs4Subjects shared task (Subtask 2) at GermEval-2025. The task required creating subject predictions for bibliographic records using large language models, with a special focus on computational efficiency. Our system, based on the Annif automated subject indexing toolkit, refines our previous system from the first LLMs4Subjects shared task, which produced excellent results. We further improved the system by using many small and efficient language mo…

physics_collab: Multilayer physicist collaborations (2015)
Two multiplex networks of coauthorships among the Pierre Auger Collaboration of physicists (2010-2012) and among researchers who have posted preprints on arXiv.org (all papers up to May 2014). Layers represent different categories of publication, and an edge's weight indicates the number of reports written by the authors. These layers are one-mode projections from the underlying author-paper bipartite network.
This n…

Vision encoders should be image size agnostic and task driven
Nedyalko Prisadnikov, Danda Pani Paudel, Yuqian Fu, Luc Van Gool
https://arxiv.org/abs/2508.16317 https://

Vision encoders should be image size agnostic and task driven
This position paper argues that the next generation of vision encoders should be image size agnostic and task driven. The source of our inspiration is biological. Not a structural aspect of biological vision, but a behavioral trait -- efficiency. We focus on a couple of ways in which vision in nature is efficient, but modern vision encoders not. We -- humans and animals -- deal with vast quantities of visual data, and need to be smart where we focus our limited energy -- it depends on the task.…

A XAI-based Framework for Frequency Subband Characterization of Cough Spectrograms in Chronic Respiratory Disease
Patricia Amado-Caballero, Luis M. San-Jos\'e-Revuelta, Xinheng Wang, Jos\'e Ram\'on Garmendia-Leiza, Carlos Alberola-L\'opez, Pablo Casaseca-de-la-Higuera
https://arxiv.org/abs/2508.16237

A XAI-based Framework for Frequency Subband Characterization of Cough Spectrograms in Chronic Respiratory Disease
This paper presents an explainable artificial intelligence (XAI)-based framework for the spectral analysis of cough sounds associated with chronic respiratory diseases, with a particular focus on Chronic Obstructive Pulmonary Disease (COPD). A Convolutional Neural Network (CNN) is trained on time-frequency representations of cough signals, and occlusion maps are used to identify diagnostically relevant regions within the spectrograms. These highlighted areas are subsequently decomposed into fiv…

Emergent Cognitive Convergence via Implementation: A Structured Loop Reflecting Four Theories of Mind (A Position Paper)
Myung Ho Kim
https://arxiv.org/abs/2507.16184

Emergent Cognitive Convergence via Implementation: A Structured Loop Reflecting Four Theories of Mind (A Position Paper)
We report the discovery of a structural convergence across four influential theories of mind: Kahneman's dual-system theory, Friston's predictive processing, Minsky's society of mind, and Clark's extended mind-emerging unintentionally within a practical AI agent architecture called Agentic Flow. Designed to address limitations in large language models (LLMs), Agentic Flow comprises five interdependent modules such as Retrieval, Cognition, Control, Memory, and Action arranged in a recurrent cogn…

SINAI at eRisk@CLEF 2025: Transformer-Based and Conversational Strategies for Depression Detection
Alba Maria Marmol-Romero, Manuel Garcia-Vega, Miguel Angel Garcia-Cumbreras, Arturo Montejo-Raez
https://arxiv.org/abs/2509.19861

SINAI at eRisk@CLEF 2025: Transformer-Based and Conversational Strategies for Depression Detection
This paper describes the participation of the SINAI-UJA team in the eRisk@CLEF 2025 lab. Specifically, we addressed two of the proposed tasks: (i) Task 2: Contextualized Early Detection of Depression, and (ii) Pilot Task: Conversational Depression Detection via LLMs. Our approach for Task 2 combines an extensive preprocessing pipeline with the use of several transformer-based models, such as RoBERTa Base or MentalRoBERTA Large, to capture the contextual and sequential nature of multi-user conve…

ColorBlindnessEval: Can Vision-Language Models Pass Color Blindness Tests?
Zijian Ling, Han Zhang, Yazhuo Zhou, Jiahao Cui
https://arxiv.org/abs/2509.19070 https://

ColorBlindnessEval: Can Vision-Language Models Pass Color Blindness Tests?
This paper presents ColorBlindnessEval, a novel benchmark designed to evaluate the robustness of Vision-Language Models (VLMs) in visually adversarial scenarios inspired by the Ishihara color blindness test. Our dataset comprises 500 Ishihara-like images featuring numbers from 0 to 99 with varying color combinations, challenging VLMs to accurately recognize numerical information embedded in complex visual patterns. We assess 9 VLMs using Yes/No and open-ended prompts and compare their performan…

GLARE: Agentic Reasoning for Legal Judgment Prediction
Xinyu Yang, Chenlong Deng, Zhicheng Dou
https://arxiv.org/abs/2508.16383 https://arxiv.org/pdf/2508.…

GLARE: Agentic Reasoning for Legal Judgment Prediction
Legal judgment prediction (LJP) has become increasingly important in the legal field. In this paper, we identify that existing large language models (LLMs) have significant problems of insufficient reasoning due to a lack of legal knowledge. Therefore, we introduce GLARE, an agentic legal reasoning framework that dynamically acquires key legal knowledge by invoking different modules, thereby improving the breadth and depth of reasoning. Experiments conducted on the real-world dataset verify the…

PyG 2.0: Scalable Learning on Real World Graphs
Matthias Fey, Jinu Sunil, Akihiro Nitta, Rishi Puri, Manan Shah, Bla\v{z} Stojanovi\v{c}, Ramona Bendias, Alexandria Barghi, Vid Kocijan, Zecheng Zhang, Xinwei He, Jan Eric Lenssen, Jure Leskovec
https://arxiv.org/abs/2507.16991

PyG 2.0: Scalable Learning on Real World Graphs
PyG (PyTorch Geometric) has evolved significantly since its initial release, establishing itself as a leading framework for Graph Neural Networks. In this paper, we present Pyg 2.0 (and its subsequent minor versions), a comprehensive update that introduces substantial improvements in scalability and real-world application capabilities. We detail the framework's enhanced architecture, including support for heterogeneous and temporal graphs, scalable feature/graph stores, and various optimization…

Hybrid Tokenization Strategy for DNA Language Model using Byte Pair Encoding and K-MER Methods
Ganesh Sapkota, Md Hasibur Rahman
https://arxiv.org/abs/2507.18570 https://…

Hybrid Tokenization Strategy for DNA Language Model using Byte Pair Encoding and K-MER Methods
This paper presents a novel hybrid tokenization strategy that enhances the performance of DNA Language Models (DLMs) by combining 6-mer tokenization with Byte Pair Encoding (BPE-600). Traditional k-mer tokenization is effective at capturing local DNA sequence structures but often faces challenges, including uneven token distribution and a limited understanding of global sequence context. To address these limitations, we propose merging unique 6mer tokens with optimally selected BPE tokens gener…

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…

E.A.R.T.H.: Structuring Creative Evolution through Model Error in Generative AI
Yusen Peng, Shuhua Mao
https://arxiv.org/abs/2507.18004 https://arxiv.org/p…

E.A.R.T.H.: Structuring Creative Evolution through Model Error in Generative AI
How can AI move beyond imitation toward genuine creativity? This paper proposes the E.A.R.T.H. framework, a five-stage generative pipeline that transforms model-generated errors into creative assets through Error generation, Amplification, Refine selection, Transform, and Harness feedback. Drawing on cognitive science and generative modeling, we posit that "creative potential hides in failure" and operationalize this via structured prompts, semantic scoring, and human-in-the-loop evaluation. Im…

One Filters All: A Generalist Filter for State Estimation
Shiqi Liu, Wenhan Cao, Chang Liu, Zeyu He, Tianyi Zhang, Shengbo Eben Li
https://arxiv.org/abs/2509.20051 https://

One Filters All: A Generalist Filter for State Estimation
Estimating hidden states in dynamical systems, also known as optimal filtering, is a long-standing problem in various fields of science and engineering. In this paper, we introduce a general filtering framework, \textbf{LLM-Filter}, which leverages large language models (LLMs) for state estimation by embedding noisy observations with text prototypes. In various experiments for classical dynamical systems, we find that first, state estimation can significantly benefit from the reasoning knowledg…

System Report for CCL25-Eval Task 10: SRAG-MAV for Fine-Grained Chinese Hate Speech Recognition
Jiahao Wang, Ramen Liu, Longhui Zhang, Jing Li
https://arxiv.org/abs/2507.18580 h…

System Report for CCL25-Eval Task 10: SRAG-MAV for Fine-Grained Chinese Hate Speech Recognition
This paper presents our system for CCL25-Eval Task 10, addressing Fine-Grained Chinese Hate Speech Recognition (FGCHSR). We propose a novel SRAG-MAV framework that synergistically integrates task reformulation(TR), Self-Retrieval-Augmented Generation (SRAG), and Multi-Round Accumulative Voting (MAV). Our method reformulates the quadruplet extraction task into triplet extraction, uses dynamic retrieval from the training set to create contextual prompts, and applies multi-round inference with vot…

VIMD: Monocular Visual-Inertial Motion and Depth Estimation
Saimouli Katragadda, Guoquan Huang
https://arxiv.org/abs/2509.19713 https://arxiv.org/pdf/2509.…

VIMD: Monocular Visual-Inertial Motion and Depth Estimation
Accurate and efficient dense metric depth estimation is crucial for 3D visual perception in robotics and XR. In this paper, we develop a monocular visual-inertial motion and depth (VIMD) learning framework to estimate dense metric depth by leveraging accurate and efficient MSCKF-based monocular visual-inertial motion tracking. At the core the proposed VIMD is to exploit multi-view information to iteratively refine per-pixel scale, instead of globally fitting an invariant affine model as in the …

Benchmarking the Robustness of Agentic Systems to Adversarially-Induced Harms
Jonathan N\"other, Adish Singla, Goran Radanovic
https://arxiv.org/abs/2508.16481 https://

Benchmarking the Robustness of Agentic Systems to Adversarially-Induced Harms
Ensuring the safe use of agentic systems requires a thorough understanding of the range of malicious behaviors these systems may exhibit when under attack. In this paper, we evaluate the robustness of LLM-based agentic systems against attacks that aim to elicit harmful actions from agents. To this end, we propose a novel taxonomy of harms for agentic systems and a novel benchmark, BAD-ACTS, for studying the security of agentic systems with respect to a wide range of harmful actions. BAD-ACTS co…

Advances in Large Language Models for Medicine
Zhiyu Kan, Wensheng Gan, Zhenlian Qi, Philip S. Yu
https://arxiv.org/abs/2509.18690 https://arxiv.org/pdf/25…

Generation of Synthetic Clinical Text: A Systematic Review
Basel Alshaikhdeeb, Ahmed Abdelmonem Hemedan, Soumyabrata Ghosh, Irina Balaur, Venkata Satagopam
https://arxiv.org/abs/2507.18451

Generation of Synthetic Clinical Text: A Systematic Review
Generating clinical synthetic text represents an effective solution for common clinical NLP issues like sparsity and privacy. This paper aims to conduct a systematic review on generating synthetic medical free-text by formulating quantitative analysis to three research questions concerning (i) the purpose of generation, (ii) the techniques, and (iii) the evaluation methods. We searched PubMed, ScienceDirect, Web of Science, Scopus, IEEE, Google Scholar, and arXiv databases for publications asso…

Structuring GUI Elements through Vision Language Models: Towards Action Space Generation
Yi Xu, Yesheng Zhang, jiajia Liu, Jingdong Chen
https://arxiv.org/abs/2508.16271 https:/…

Structuring GUI Elements through Vision Language Models: Towards Action Space Generation
Multimodal large language models (MLLMs) have emerged as pivotal tools in enhancing human-computer interaction. In this paper we focus on the application of MLLMs in the field of graphical user interface (GUI) elements structuring, where they assist in processing user instructions based on screen contents. Despite the promise of MLLMs, their performance in precisely generating UI element coordinates, a critical aspect of GUI understanding, is hindered by the nature of next-token prediction trai…

Cross-Cultural Transfer of Commonsense Reasoning in LLMs: Evidence from the Arab World
Saeed Almheiri, Rania Hossam, Mena Attia, Chenxi Wang, Preslav Nakov, Timothy Baldwin, Fajri Koto
https://arxiv.org/abs/2509.19265

Unsupervised Online Detection of Pipe Blockages and Leakages in Water Distribution Networks
Jin Li, Kleanthis Malialis, Stelios G. Vrachimis, Marios M. Polycarpou
https://arxiv.org/abs/2508.16336

Unsupervised Online Detection of Pipe Blockages and Leakages in Water Distribution Networks
Water Distribution Networks (WDNs), critical to public well-being and economic stability, face challenges such as pipe blockages and background leakages, exacerbated by operational constraints such as data non-stationarity and limited labeled data. This paper proposes an unsupervised, online learning framework that aims to detect two types of faults in WDNs: pipe blockages, modeled as collective anomalies, and background leakages, modeled as concept drift. Our approach combines a Long Short-Ter…

Hybrid Annotation for Propaganda Detection: Integrating LLM Pre-Annotations with Human Intelligence
Ariana Sahitaj, Premtim Sahitaj, Veronika Solopova, Jiaao Li, Sebastian M\"oller, Vera Schmitt
https://arxiv.org/abs/2507.18343

Hybrid Annotation for Propaganda Detection: Integrating LLM Pre-Annotations with Human Intelligence
Propaganda detection on social media remains challenging due to task complexity and limited high-quality labeled data. This paper introduces a novel framework that combines human expertise with Large Language Model (LLM) assistance to improve both annotation consistency and scalability. We propose a hierarchical taxonomy that organizes 14 fine-grained propaganda techniques into three broader categories, conduct a human annotation study on the HQP dataset that reveals low inter-annotator agreeme…

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…

Confounded Causal Imitation Learning with Instrumental Variables
Yan Zeng, Shenglan Nie, Feng Xie, Libo Huang, Peng Wu, Zhi Geng
https://arxiv.org/abs/2507.17309

Confounded Causal Imitation Learning with Instrumental Variables
Imitation learning from demonstrations usually suffers from the confounding effects of unmeasured variables (i.e., unmeasured confounders) on the states and actions. If ignoring them, a biased estimation of the policy would be entailed. To break up this confounding gap, in this paper, we take the best of the strong power of instrumental variables (IV) and propose a Confounded Causal Imitation Learning (C2L) model. This model accommodates confounders that influence actions across multiple timest…

The Mediomatix Corpus: Parallel Data for Romansh Idioms via Comparable Schoolbooks
Zachary Hopton, Jannis Vamvas, Andrin B\"uchler, Anna Rutkiewicz, Rico Cathomas, Rico Sennrich
https://arxiv.org/abs/2508.16371

The Mediomatix Corpus: Parallel Data for Romansh Idioms via Comparable Schoolbooks
The five idioms (i.e., varieties) of the Romansh language are largely standardized and are taught in the schools of the respective communities in Switzerland. In this paper, we present the first parallel corpus of Romansh idioms. The corpus is based on 291 schoolbook volumes, which are comparable in content for the five idioms. We use automatic alignment methods to extract 207k multi-parallel segments from the books, with more than 2M tokens in total. A small-scale human evaluation confirms tha…

Adaptive Guidance Semantically Enhanced via Multimodal LLM for Edge-Cloud Object Detection
Yunqing Hu, Zheming Yang, Chang Zhao, Wen Ji
https://arxiv.org/abs/2509.19875 https://…

Adaptive Guidance Semantically Enhanced via Multimodal LLM for Edge-Cloud Object Detection
Traditional object detection methods face performance degradation challenges in complex scenarios such as low-light conditions and heavy occlusions due to a lack of high-level semantic understanding. To address this, this paper proposes an adaptive guidance-based semantic enhancement edge-cloud collaborative object detection method leveraging Multimodal Large Language Models (MLLM), achieving an effective balance between accuracy and efficiency. Specifically, the method first employs instructio…

Predictive Coding-based Deep Neural Network Fine-tuning for Computationally Efficient Domain Adaptation
Matteo Cardoni, Sam Leroux
https://arxiv.org/abs/2509.20269 https://

Predictive Coding-based Deep Neural Network Fine-tuning for Computationally Efficient Domain Adaptation
As deep neural networks are increasingly deployed in dynamic, real-world environments, relying on a single static model is often insufficient. Changes in input data distributions caused by sensor drift or lighting variations necessitate continual model adaptation. In this paper, we propose a hybrid training methodology that enables efficient on-device domain adaptation by combining the strengths of Backpropagation and Predictive Coding. The method begins with a deep neural network trained offli…

Effective Multi-Task Learning for Biomedical Named Entity Recognition
Jo\~ao Ruano, Gon\c{c}alo M. Correia, Leonor Barreiros, Afonso Mendes
https://arxiv.org/abs/2507.18542 http…

Effective Multi-Task Learning for Biomedical Named Entity Recognition
Biomedical Named Entity Recognition presents significant challenges due to the complexity of biomedical terminology and inconsistencies in annotation across datasets. This paper introduces SRU-NER (Slot-based Recurrent Unit NER), a novel approach designed to handle nested named entities while integrating multiple datasets through an effective multi-task learning strategy. SRU-NER mitigates annotation gaps by dynamically adjusting loss computation to avoid penalizing predictions of entity types …

Long Story Short: Disentangling Compositionality and Long-Caption Understanding in VLMs
Israfel Salazar, Desmond Elliott, Yova Kementchedjhieva
https://arxiv.org/abs/2509.19207 …

DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD
Xianbiao Qi, Marco Chen, Wenjie Xiao, Jiaquan Ye, Yelin He, Chun-Guang Li, Zhouchen Lin
https://arxiv.org/abs/2507.17501

DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD
Transformers have become the de facto backbone of modern deep learning, yet their training typically demands an advanced optimizer with adaptive learning rate like AdamW, rather than a momentum SGDW (mSGDW). Previous works show that it is mainly due to a heavy-tailed distribution of the gradients. In this paper, we introduce a Deeply Normalized Transformer (DNT), which is meticulously engineered to overcome this limitation enabling seamless training with vanilla mSGDW while yielding comparable …

Evolutionary Feature-wise Thresholding for Binary Representation of NLP Embeddings
Soumen Sinha, Shahryar Rahnamayan, Azam Asilian Bidgoli
https://arxiv.org/abs/2507.17025

Evolutionary Feature-wise Thresholding for Binary Representation of NLP Embeddings
Efficient text embedding is crucial for large-scale natural language processing (NLP) applications, where storage and computational efficiency are key concerns. In this paper, we explore how using binary representations (barcodes) instead of real-valued features can be used for NLP embeddings derived from machine learning models such as BERT. Thresholding is a common method for converting continuous embeddings into binary representations, often using a fixed threshold across all features. We pr…

Generative Adversarial Networks Applied for Privacy Preservation in Biometric-Based Authentication and Identification
Lubos Mjachky, Ivan Homoliak
https://arxiv.org/abs/2509.20024

Generative Adversarial Networks Applied for Privacy Preservation in Biometric-Based Authentication and Identification
Biometric-based authentication systems are getting broadly adopted in many areas. However, these systems do not allow participating users to influence the way their data is used. Furthermore, the data may leak and can be misused without the users' knowledge. In this paper, we propose a new authentication method that preserves the privacy of individuals and is based on a generative adversarial network (GAN). Concretely, we suggest using the GAN for translating images of faces to a visually priva…

Decentralized Federated Learning of Probabilistic Generative Classifiers
Aritz P\'erez, Carlos Echegoyen, Guzm\'an Santaf\'e
https://arxiv.org/abs/2507.17285

Decentralized Federated Learning of Probabilistic Generative Classifiers
Federated learning is a paradigm of increasing relevance in real world applications, aimed at building a global model across a network of heterogeneous users without requiring the sharing of private data. We focus on model learning over decentralized architectures, where users collaborate directly to update the global model without relying on a central server. In this context, the current paper proposes a novel approach to collaboratively learn probabilistic generative classifiers with a parame…

Beyond Human-prompting: Adaptive Prompt Tuning with Semantic Alignment for Anomaly Detection
Pi-Wei Chen, Jerry Chun-Wei Lin, Wei-Han Chen, Jia Ji, Zih-Ching Chen, Feng-Hao Yeh, Chao-Chun Chen
https://arxiv.org/abs/2508.16157

Beyond Human-prompting: Adaptive Prompt Tuning with Semantic Alignment for Anomaly Detection
Pre-trained Vision-Language Models (VLMs) have recently shown promise in detecting anomalies. However, previous approaches are fundamentally limited by their reliance on human-designed prompts and the lack of accessible anomaly samples, leading to significant gaps in context-specific anomaly understanding. In this paper, we propose \textbf{A}daptive \textbf{P}rompt \textbf{T}uning with semantic alignment for anomaly detection (APT), a groundbreaking prior knowledge-free, few-shot framework and …

Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models
Zhifang Zhang, Jiahan Zhang, Shengjie Zhou, Qi Wei, Shuo He, Feng Liu, Lei Feng
https://arxiv.org/abs/2509.19994

Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models
Multimodal pre-trained models (e.g., ImageBind), which align distinct data modalities into a shared embedding space, have shown remarkable success across downstream tasks. However, their increasing adoption raises serious security concerns, especially regarding targeted adversarial attacks. In this paper, we show that existing targeted adversarial attacks on multimodal pre-trained models still have limitations in two aspects: generalizability and undetectability. Specifically, the crafted targe…

U-Mamba2-SSL for Semi-Supervised Tooth and Pulp Segmentation in CBCT
Zhi Qin Tan, Xiatian Zhu, Owen Addison, Yunpeng Li
https://arxiv.org/abs/2509.20154 https://

U-Mamba2-SSL for Semi-Supervised Tooth and Pulp Segmentation in CBCT
Accurate segmentation of teeth and pulp in Cone-Beam Computed Tomography (CBCT) is vital for clinical applications like treatment planning and diagnosis. However, this process requires extensive expertise and is exceptionally time-consuming, highlighting the critical need for automated algorithms that can effectively utilize unlabeled data. In this paper, we propose U-Mamba2-SSL, a novel semi-supervised learning framework that builds on the U-Mamba2 model and employs a multi-stage training stra…

Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruning
Sanish Suwal, Shaurya Garg, Dipkamal Bhusal, Michael Clifford, Nidhi Rastogi
https://arxiv.org/abs/2509.20148

Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruning
Connected and autonomous vehicles continue to heavily rely on AI systems, where transparency and security are critical for trust and operational safety. Post-hoc explanations provide transparency to these black-box like AI models but the quality and reliability of these explanations is often questioned due to inconsistencies and lack of faithfulness in representing model decisions. This paper systematically examines the impact of three widely used training approaches, namely natural training, a…