Tootfinder

Colts' Shane Steichen self-evaluating play calls after collapse to Chiefs https://www.nfl.com/news/colts-shane-steichen-self-evaluating-play-calls-after-collapse-to-chiefs

A Characterization of Triangle-Free Cyclic Graphs With Self-Loops Of Rank 3
Johnny Lim
https://arxiv.org/abs/2509.20158 https://arxiv.org/pdf/2509.20158

A Characterization of Triangle-Free Cyclic Graphs With Self-Loops Of Rank 3
Let $G_S$ be a self-loop graph as the graph obtained by attaching a self-loop at every vertex in $S \subseteq V(G)$ of a simple graph $G.$ If $G=C_n$ is the cycle graphs of order $n$ and $S \neq \emptyset,$ we show that there are no rank 3 self-loop graphs $(C_n)_S$ for $n\geq 5.$ As a consequence, we determine and construct all possible rank 3 triangle-free self-loop cyclic graph of order at least 4 from $(C_4)_S$ via graph join operations. This provides a partial solution to the characterizat…

Nonlocal Games and Self-tests in the Presence of Noise
Honghao Fu, Minglong Qin, Haochen Xu, Penghui Yao
https://arxiv.org/abs/2509.20350 https://arxiv.org…

Nonlocal Games and Self-tests in the Presence of Noise
Self-testing is a key characteristic of certain nonlocal games, which allow one to uniquely determine the underlying quantum state and measurement operators used by the players, based solely on their observed input-output correlations [MY04]. Motivated by the limitations of current quantum devices, we study self-testing in the high-noise regime, where the two players are restricted to sharing many copies of a noisy entangled state with an arbitrary constant noise rate. In this setting, many exi…

Looks like blockchains have finally found a serious use case, just maybe not the one predicted/hoped for...
This article is about #Glassworm, the latest major exploit in JavaScript-land, targetting VSCode and using #Solana as command infrastructure and Google Calendar events as backup. It'…

GlassWorm: First Self-Propagating Worm Using Invisible Code Hits OpenVSX Marketplace | Koi Blog

Estimating the Self-Consistency of LLMs
Robert Nowak
https://arxiv.org/abs/2509.19489 https://arxiv.org/pdf/2509.19489

Estimating the Self-Consistency of LLMs
Systems often repeat the same prompt to large language models (LLMs) and aggregate responses to improve reliability. This short note analyzes an estimator of the self-consistency of LLMs and the tradeoffs it induces under a fixed compute budget $B=mn$, where $m$ is the number of prompts sampled from the task distribution and $n$ is the number of repeated LLM calls per prompt; the resulting analysis favors a rough split $m,n\propto\sqrt{B}$.

Dual-Language General-Purpose Self-Hosted Visual Language and new Textual Programming Language for Applications
Mahmoud Samir Fayed
https://arxiv.org/abs/2509.20426 https://

Dual-Language General-Purpose Self-Hosted Visual Language and new Textual Programming Language for Applications
Most visual programming languages (VPLs) are domain-specific, with few general-purpose VPLs like Programming Without Coding Technology (PWCT). These general-purpose VPLs are developed using textual programming languages and improving them requires textual programming. In this thesis, we designed and developed PWCT2, a dual-language (Arabic/English), general-purpose, self-hosting visual programming language. Before doing so, we specifically designed a textual programming language called Ring for…

Self-evolved Imitation Learning in Simulated World
Yifan Ye, Jun Cen, Jing Chen, Zhihe Lu
https://arxiv.org/abs/2509.19460 https://arxiv.org/pdf/2509.19460…

Self-evolved Imitation Learning in Simulated World
Imitation learning has been a trend recently, yet training a generalist agent across multiple tasks still requires large-scale expert demonstrations, which are costly and labor-intensive to collect. To address the challenge of limited supervision, we propose Self-Evolved Imitation Learning (SEIL), a framework that progressively improves a few-shot model through simulator interactions. The model first attempts tasksin the simulator, from which successful trajectories are collected as new demonst…

Online Adaptation via Dual-Stage Alignment and Self-Supervision for Fast-Calibration Brain-Computer Interfaces
Sheng-Bin Duan, Jian-Long Hao, Tian-Yu Xiang, Xiao-Hu Zhou, Mei-Jiang Gui, Xiao-Liang Xie, Shi-Qi Liu, Zeng-Guang Hou
https://arxiv.org/abs/2509.19403

Online Adaptation via Dual-Stage Alignment and Self-Supervision for Fast-Calibration Brain-Computer Interfaces
Individual differences in brain activity hinder the online application of electroencephalogram (EEG)-based brain computer interface (BCI) systems. To overcome this limitation, this study proposes an online adaptation algorithm for unseen subjects via dual-stage alignment and self-supervision. The alignment process begins by applying Euclidean alignment in the EEG data space and then updates batch normalization statistics in the representation space. Moreover, a self-supervised loss is designed …

WPP rolls out WPP Open Pro, a self-serve version of WPP Open, its AI-powered marketing platform that helps marketers build campaigns and develop creative assets (Kendra Barnett/Adweek)
https://www.adweek.com/agencies/wpp-launch…

WPP Launches a Self-Serve Version of Its AI Platform To Court Small Biz Dollars
British advertising titan WPP is rolling out a self-serve version of WPP Open, its AI marketing platform. 

EvoMail: Self-Evolving Cognitive Agents for Adaptive Spam and Phishing Email Defense
Wei Huang, De-Tian Chu, Lin-Yuan Bai, Wei Kang, Hai-Tao Zhang, Bo Li, Zhi-Mo Han, Jing Ge, Hai-Feng Lin
https://arxiv.org/abs/2509.21129

EvoMail: Self-Evolving Cognitive Agents for Adaptive Spam and Phishing Email Defense
Modern email spam and phishing attacks have evolved far beyond keyword blacklists or simple heuristics. Adversaries now craft multi-modal campaigns that combine natural-language text with obfuscated URLs, forged headers, and malicious attachments, adapting their strategies within days to bypass filters. Traditional spam detection systems, which rely on static rules or single-modality models, struggle to integrate heterogeneous signals or to continuously adapt, leading to rapid performance degra…

Instruction-tuned Self-Questioning Framework for Multimodal Reasoning
You-Won Jang, Yu-Jung Heo, Jaeseok Kim, Minsu Lee, Du-Seong Chang, Byoung-Tak Zhang
https://arxiv.org/abs/2509.21251

Instruction-tuned Self-Questioning Framework for Multimodal Reasoning
The field of vision-language understanding has been actively researched in recent years, thanks to the development of Large Language Models~(LLMs). However, it still needs help with problems requiring multi-step reasoning, even for very simple questions. Recent studies adopt LLMs to tackle this problem by iteratively generating sub-questions and answers. However, there are disadvantages such as 1) the fine-grained visual contents of images are not available using LLMs that cannot read visual in…

👾 Unsupervised Instance Segmentation with Superpixels
#cs #arxiv

Unsupervised Instance Segmentation with Superpixels
Instance segmentation is essential for numerous computer vision applications, including robotics, human-computer interaction, and autonomous driving. Currently, popular models bring impressive performance in instance segmentation by training with a large number of human annotations, which are costly to collect. For this reason, we present a new framework that efficiently and effectively segments objects without the need for human annotations. Firstly, a MultiCut algorithm is applied to self-sup…

Zijn hier debatten over geweest? Die heb ik dan gemist
https://www.autoblog.nl/nieuws/nederland-eerste-eu-land-waar-tesla-full-self-driving-legaal-is-7182233

Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation
Wei-Hsiang Lin, Sheng-Lun Wei, Hen-Hsen Huang, Hsin-Hsi Chen
https://arxiv.org/abs/2509.19880 https://

Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation
LLM-as-Judge frameworks are increasingly popular for AI evaluation, yet research findings on the relationship between models' generation and judgment abilities remain inconsistent. We investigate this relationship through systematic dataset- and instance-level analyses across 11 models and 21 diverse tasks. Despite both capabilities relying on the same underlying knowledge, our analyses reveal they are only weakly correlated, primarily due to LLMs' sensitivity to the responses being judged. To …

Anderson self-localization of light in pair plasmas
Maxim Lyutikov (Purdue University), Victor Gurarie (University of Colorado)
https://arxiv.org/abs/2509.20594 https://

Anderson self-localization of light in pair plasmas
We demonstrate that in pair plasma weakly nonlinear electromagnetic waves, $a_0 \leq 1$, experience Anderson self-localization. The beat between the driver and a back-scattered wave creates random, charge-neutral, large density fluctuations $δρ/ρ\gg 1$, and corresponding random fluctuations of the dielectric permittivity $ε$. Propagating in quasi-1D, waves in a medium with spatially random, time-varying, self-created fluctuations of dielectric permeability experience localization. Anderson …

Frequency-domain self-force calculations using Gegenbauer reconstruction
Christopher Whittall, Leor Barack, Oliver Long
https://arxiv.org/abs/2509.19439 https://

Frequency-domain self-force calculations using Gegenbauer reconstruction
We investigate the use of the Gegenbauer procedure for time-domain reconstruction in frequency-domain calculations of the self-force on a particle orbiting a black hole. The conventional technique relies on the so-called method of extended homogeneous solutions (EHS), which circumvents the Gibbs phenomenon that would otherwise hamper the reconstruction of the particle's field from frequency modes. Unfortunately, the effectiveness of EHS reconstruction deteriorates rapidly with increasing orbita…

Sonnet 038 - XXXVIII
How can my muse want subject to invent,
While thou dost breathe, that pour'st into my verse
Thine own sweet argument, too excellent
For every vulgar paper to rehearse?
O! give thy self the thanks, if aught in me
Worthy perusal stand against thy sight;
For who's so dumb that cannot write to thee,
When thou thy self dost give invention light?
Be thou the tenth Muse, ten times more in worth
Than those old n…

⚙️ Flexible configuration options including temperature control, context windows, timeouts & customizable prompts for fine-tuned responses
🌐 https://tighten.com/insights/build-private-self-hosted-ai-applications-with-ollama-and-laravel/

DARKSKIES: A suite of super-sampled zoom-in simulations of galaxy clusters with self-interacting dark matter
David Harvey, Yves Revaz, Matthieu Schaller, Aurel Schneider, Ethan Tregidga, Felix Vecchi
https://arxiv.org/abs/2509.19842

DARKSKIES: A suite of super-sampled zoom-in simulations of galaxy clusters with self-interacting dark matter
We present the "DARKSKIES" suite of one hundred, zoom-in hydrodynamic simulations of massive ($M_{200}>5\times10^{14}{\rm M}_\odot)$ galaxy clusters with self-interacting dark matter (SIDM). We super-sample the simulations such that $m_{\rm DM}/m_{\rm gas}\sim0.1$, enabling us to simulate a dark matter particle mass of $m=0.68\times10^{8}M_\odot$ an order of magnitude faster, whilst exploring SIDM in the core of clusters at extremely high resolution. We calibrate the baryonic feedback to produc…

Tesla influencers just tried Elon Musk’s coast-to-coast self-driving challenge from 2017, crashed before 60 miles | Electrek https://electrek.co/2025/09/21/tesla-influencers-tried-elon-musk-coast-to-coast-self-driving-crashed-be…

please enjoy: my Wasm-hosted, Wasm-targeting build of Clang/Clang /LLD: a self-contained, 25 MiB (gzipped) pure function
https://www.npmjs.com/package/@yowasp/clang

Dynamic self-assembly of active particle systems controlled by light fields
Sihang Guo, Guangyu Yang, Guoqing Meng, Yingying Wang, Junxing Pan, Jinjun Zhang
https://arxiv.org/abs/2509.19971

Dynamic self-assembly of active particle systems controlled by light fields
Active particle systems are a class of non-equilibrium systems composed of self-propelled Brownian particles; through interactions between particles within the system, a variety of intriguing collective behaviors can emerge. Based on Brownian dynamics simulations, this study investigates the formation and transition mechanisms of ordered structures in active particle systems regulated by light fields. The study reveals that, under light field regulation, active particles undergo large-scale pha…

Here's my latest Human Meme podcast about self-awareness and consciousness:
https://humanmeme.com/the-mirror-and-the-machine-a-meditation-on-consciousness-and-self-awareness

"I actually think acting is a form of self-hypnosis. You have to be hyper, hyper aware of what's going on around you. You have to know where the lens is, what the shot is, and where you're moving. And then you have to trick yourself into an emotional state where you believe this stuff is actually happening."
—Julianne Moore
#acting

Enhanced White-Light Emission from Self-Trapped Excitons in Antimony and Bismuth Halides through Structural Design
Philip Klement, Lukas G\"umbel, Meng Yang, Jan-Heinrich Littmann, Tatsuhiko Ohto, Hirokazu Tada, Sangam Chatterjee, Johanna Heine
https://arxiv.org/abs/2509.20087

Enhanced White-Light Emission from Self-Trapped Excitons in Antimony and Bismuth Halides through Structural Design
Lead halide perovskites have catalyzed the rise of main-group metal halide materials as promising candidates for next-generation optoelectronics, including solar cells, light-emitting diodes, lasers, sensors, and photocatalysts. Among these, effi-cient light-emission arises from self-trapped excitons, wherein excited states induce transient lattice distortions that localize excitons. However, the complex interplay of factors, such as lattice distortions, lattice softness, and electron-phonon co…

Filing: OpenAI denies liability in a suit alleging ChatGPT gave info about suicide methods to a 16-year-old who died by suicide, arguing he misused the chatbot (Angela Yang/NBC News)
https://www.nbcnews.com/tech/tech-news/ope

OpenAI denies allegations that ChatGPT is to blame for a teenager's suicide
Warning: This article includes descriptions of self-harm

petster: Pet owner social networks (2016)
A network of friendships among users on catster.com and dogster.com. Node represents a user, and edge exists if two users self-identify as friends on either website. The dogster/catster network also contains family links between dog and dog, cat and cat, as well as dog and cat.
This network has 623766 nodes and 15699276 edges.
Tags: Social, Online, Unweighted

And yet another post talking about mapping my bike rides... this time looking at the very interesting wanderer which is open source and uses ActivityPub!
https://rasterweb.net/raster/2025/09/26/mapping-bike-rides-part-iv/

Mapping Bike Rides (Part IV)
An open source, self-hosted route mapping solution...

"AI will take over the world."

This can mean two things:

1. "AI" will become self-aware and decide to control humans.

2. Gen-slop and deepfakes will dominate available content, leading to deskilling, human abuse and eco-harm.

The first idea has no basis, but it's fair to worry about the second.

Estimating the Self-Consistency of LLMs
Robert Nowak
https://arxiv.org/abs/2509.19489 https://arxiv.org/pdf/2509.19489

Estimating the Self-Consistency of LLMs
Systems often repeat the same prompt to large language models (LLMs) and aggregate responses to improve reliability. This short note analyzes an estimator of the self-consistency of LLMs and the tradeoffs it induces under a fixed compute budget $B=mn$, where $m$ is the number of prompts sampled from the task distribution and $n$ is the number of repeated LLM calls per prompt; the resulting analysis favors a rough split $m,n\propto\sqrt{B}$.

Self-embeddings of homogeneous self-similar sets generated by three maps
Zhiqiang Wang
https://arxiv.org/abs/2509.18202 https://arxiv.org/pdf/2509.18202

Self-embeddings of homogeneous self-similar sets generated by three maps
For $0< ρ< 1/3$ and $ρ\le λ\le 1-2ρ$, let $E$ be the self-similar set generated by the iterated function system $$Φ= \big\{ φ_1(x) = ρx ,\; φ_2(x) = ρx + λ, \; φ_3(x) = ρx + 1- ρ\big\}.$$ All contractive similitudes $f$ with $f(E) \subset E$ are characterized: one can find $i_1, i_2, \ldots, i_n \in \{1,2,3\}$ such that \[ f(E)=φ_{i_1} \circ φ_{i_2} \circ \cdots \circ φ_{i_n} (E). \]

A self-consistent numerical model of internal wave-induced mean flow oscillations in polar geometry
Florentin Daniel, Daniel Lecoanet
https://arxiv.org/abs/2509.19482 https://…

A self-consistent numerical model of internal wave-induced mean flow oscillations in polar geometry
The Earth's Quasi-Biennial Oscillation (QBO) is a natural example of wave-mean flow interaction and corresponds to the alternating directions of winds in the equatorial stratosphere. It is due to internal gravity waves (IGW) generated in the underlying convective troposphere. In stars, a similar situation is predicted to occur, with the interaction of a stably-stratified radiative zone and a convective zone. In this context, we investigate the dynamics of this reversing mean flow by modelling a…

The A.C.L.U. challenge accuses the current administration of housing small numbers of men at Guantšnamo, rather then in U.S. facilities,
as part of a messaging strategy
“to frighten immigrants, deter future migration, induce self-deportation and coerce people in detention to give up claims against removal and accept deportation elsewhere.”
It has called conditions inhumane, and the detainees’ access to legal counsel inadequate.
The administration contends that the de…

Trump Official Says U.S. Can House Migrants at All of Its Overseas Bases
A Justice Department lawyer made the claim in response to a challenge to the administration’s use of the base at Guantánamo Bay, Cuba, to hold detainees designated for deportation.

An overview of neural architectures for self-supervised audio representation learning from masked spectrograms
Sarthak Yadav, Sergios Theodoridis, Zheng-Hua Tan
https://arxiv.org/abs/2509.18691

Finishing out the week with some fuzzed out heavy stoner sludge via Las Vegas' PLAGUE DOCTOR from their new self-titled LP. Really digging this one. Riffs for days
https://plaguedoctorlv.bandcamp.com/album/plague-doctor

Damn.. the new #Cloudron UI revamp is sick 😲 loving it! https://cloudron.io

Cloudron - Complete solution for self-hosting
Cloudron the best platform for self-hosting web apps.

Anomalous Quantum Relaxation in the Infinite Temperature Hubbard Chain
Catalin Pascu Moca, Bal\'azs D\'ora
https://arxiv.org/abs/2509.20759 https://

Anomalous Quantum Relaxation in the Infinite Temperature Hubbard Chain
The self-energy encodes the fundamental lifetime of quasiparticle excitations. In one dimension, it is known to display anomalous behavior at zero temperature for interacting fermions, reflecting the breakdown of Fermi-liquid theory. Here we show that the self-energy is also anomalous in the infinite temperature Hubbard chain, where thermal fluctuations are maximal. Focusing on the second order ring diagram, we find that the imaginary part of the self-energy diverges non-perturbatively: as a po…

Hey fedi. I have a secret I've been keeping. I've lived in the dark for so long, hiding myself in self-loathing, and I can't do it anymore.
I am sexually fixated on Blackwater's kamikaze drones. The phallic nature of the tip its wings curve to, the build in intensity as they approach their target. The climax. Oh the climax. Kaboom, baby.
My first encounter with one was how it obliterated my family's home when I was four, by accident, en route to the president…

GlassWorm: First Self-Propagating Worm Using Invisible Code Hits OpenVSX Marketplace
A month after Shai Hulud became the first self-propagating worm in the npm ecosystem, we just discovered the world's first worm targeting VS Code extensions on OpenVSX marketplace.
🐛 https://www.

GlassWorm: First Self-Propagating Worm Using Invisible Code Hits OpenVSX Marketplace | Koi Blog

These death threats by the maga-klan create a real fear.
These kinds of statements especially as they are coupled to an increasing tendency on the part of our so-called government to move in the direction of these statements, creates a plausible defense that could be raised by the killer of Kirk and also by Luigi:
Self defense and defense of others.
Tyler Robinson and Luigi M. could well, and plausibly assert, that they were acting to defend you and me.
It's a def…

Judicious partitions for restricted self-sumsets in cyclic groups
Keane Maverick Irawan
https://arxiv.org/abs/2509.20568 https://arxiv.org/pdf/2509.20568…

Judicious partitions for restricted self-sumsets in cyclic groups
We study the minimax problem for restricted two-fold self-sumsets in $k$-colorings of $\mathbb{Z}_n$. For primes $p$ with $2\le k\le p$ we determine the exact minimum $\max\{0,\,2\lceil p/k\rceil-3\}$. For general $n$ (with $m=\lceil n/k\rceil$) we bound the optimum between a size term $\min\{p(n),\,2m-3\}$ and a periodicity term $f\big(n/q(n,k)\big)$, and show these bounds are tight when $2m-3\le p(n)$ or $f\big(n/q(n,k)\big)\le \min\{p(n),\,2m-3\}$. We further prove a stability inequality and…

I have long lived by the principles of community, voluntary cooperation, and decentralized member-driven governance. Before knowing the term anarcho-syndicalism, I often imagined unions as truly democratic organizations, where delegates are elected and revocable at any moment, rooted in direct workers’ self-management and direct participation in controlling economic and social life, independent from both state authority and capitalist ownership.
Discovering anarcho-syndicalism about a…

SAGAbg III: Environmental Stellar Mass Functions, Self-Quenching, and the Stellar-to-Halo Mass Relation in the Dwarf Galaxy Regime
Erin Kado-Fong, Yao-Yuan Mao, Yasmeen Asali, Marla Geha, Risa H. Wechsler, Mithi A. C. de los Reyes, Yunchong Wang, Ethan O. Nadler, Nitya Kallivayalil, Erik J. Tollerud, Benjamin Weiner
https://arxiv.org/abs/2…

SAGAbg III: Environmental Stellar Mass Functions, Self-Quenching, and the Stellar-to-Halo Mass Relation in the Dwarf Galaxy Regime
Recent efforts have extended our view of the number and properties of satellite galaxies beyond the Local Group firmly down to $\rm M_\star\sim 10^6 M_\odot$. A similarly complete view of the field dwarf population has lagged behind. Using the background galaxies sample from the Satellites Around Galactic Analogs (SAGA) Survey at $z<0.05$, we take inventory of the dwarf population down to $\rm M_\star \sim 5\times10^6 M_\odot$ using three metrics: the stellar mass function (SMF) as function of …

#WarOnSafety

#WarOnSafety

On self-similarity of finitely generated torsion-free nilpotent groups
Adilson Berlatto, Tulio Santos
https://arxiv.org/abs/2509.16947 https://arxiv.org/pd…

On self-similarity of finitely generated torsion-free nilpotent groups
This paper aims to investigate the self-similarity property in finitely-generated torsion-free nilpotent groups. We establish connections between geometric equivalence and self-similarity in these groups. Moreover, we show that any 2-generated torsion-free nilpotent group of class 3 has a faithful representation as a transitive self-similar group. Additionally, we provide an example of a 4-generated torsion-free nilpotent group of class 3 that does not admit such a representation. This example,…

TinyAC: Bringing Autonomic Computing Principles to Resource-Constrained Systems
Wojciech Kalka, Ruitao Xue, Kamil Faber, Aleksander Slominski, Devki Jha, Rajiv Ranjan, Tomasz Szydlo
https://arxiv.org/abs/2509.19350

TinyAC: Bringing Autonomic Computing Principles to Resource-Constrained Systems
Autonomic Computing (AC) is a promising approach for developing intelligent and adaptive self-management systems at the deep network edge. In this paper, we present the problems and challenges related to the use of AC for IoT devices. Our proposed hybrid approach bridges bottom-up intelligence (TinyML and on-device learning) and top-down guidance (LLMs) to achieve a scalable and explainable approach for developing intelligent and adaptive self-management tiny systems. Moreover, we argue that Ti…

Rough days. I believe it will get better, as there is so much talent in the squad. But it’s increasingly difficult to see the vision and that there is any tangible momentum to get there. Change can be chaotic, but the many self-inflicted injuries don’t help. Some roster decisions, e.g. Salah, Kerkez, etc. don’t seem supportable.
The failure to land Guehi mixed with the injury to Leoni have kept an out of form Konate in the lineup. Gomez? Endo?

This day 30 years ago - on Oct 26th 1995, #CPAN (the Comprehensive Perl Archive Network) was announced to the world by Jarkko Hietaniemi – CPAN's Self-Appointed Master Librarian (OOK!), to the comp.lang.perl.announce newsgroup.
Happy Birthday, CPAN!

Cowboys overcome self-inflicted wounds vs. Eagles for future-altering win, keep playoff hopes alive

https://www.cbssports.com/nfl/news/cowboys

#EroticMusings 26: How is masculinity visually expressed, highlighted, or accessorised in your work or setting?
My current roster of main characters are three cis lesbians (I plan to un-cis one of those in later books, though) and Erytan society has no other visible gender marker beyond "self-declaration".
Book three even makes the conscious decision of depiction tran…

SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent
Yandan Yang, Baoxiong Jia, Shujie Zhang, Siyuan Huang
https://arxiv.org/abs/2509.20414 ht…

SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent
Indoor scene synthesis has become increasingly important with the rise of Embodied AI, which requires 3D environments that are not only visually realistic but also physically plausible and functionally diverse. While recent approaches have advanced visual fidelity, they often remain constrained to fixed scene categories, lack sufficient object-level detail and physical consistency, and struggle to align with complex user instructions. In this work, we present SceneWeaver, a reflective agentic f…

Capitalism is just one witty ad campaign after another until we suddenly find ourselves extinct due to self-inflicted habitat loss. https://mastodon.online/@pluszysta/115428982968522861

SpellerSSL: Self-Supervised Learning with P300 Aggregation for Speller BCIs
Jiazhen Hong, Geoff Mackellar, Soheila Ghane
https://arxiv.org/abs/2509.19401 https://

SpellerSSL: Self-Supervised Learning with P300 Aggregation for Speller BCIs
Electroencephalogram (EEG)-based P300 speller brain-computer interfaces (BCIs) face three main challenges: low signal-to-noise ratio (SNR), poor generalization, and time-consuming calibration. We propose SpellerSSL, a framework that combines self-supervised learning (SSL) with P300 aggregation to address these issues. First, we introduce an aggregation strategy to enhance SNR. Second, to achieve generalization in training, we employ a customized 1D U-Net backbone and pretrain the model on both …

from my link log —
Advanced self-aware ed(1).
https://aartaka.me/advanc-ed.html
saved 2025-10-22 https://dotat.at/:/IPPZT.html

Advanced Self-Aware ed(1)
ed(1) is a versatile programming system. Yet no one talks about metaprogramming and algorithms in it. Now someone did, and that’s me!

MACD: Multi-Agent Clinical Diagnosis with Self-Learned Knowledge for LLM
Wenliang Li, Rui Yan, Xu Zhang, Li Chen, Hongji Zhu, Jing Zhao, Junjun Li, Mengru Li, Wei Cao, Zihang Jiang, Wei Wei, Kun Zhang, Shaohua Kevin Zhou
https://arxiv.org/abs/2509.20067

MACD: Multi-Agent Clinical Diagnosis with Self-Learned Knowledge for LLM
Large language models (LLMs) have demonstrated notable potential in medical applications, yet they face substantial challenges in handling complex real-world clinical diagnoses using conventional prompting methods. Current prompt engineering and multi-agent approaches typically optimize isolated inferences, neglecting the accumulation of reusable clinical experience. To address this, this study proposes a novel Multi-Agent Clinical Diagnosis (MACD) framework, which allows LLMs to self-learn cli…

"Under the new rules, online service providers will be required to assess the risk that their services could be misused for the dissemination of child sexual abuse material or for the solicitation of children."
Where do homelab and self hosting stuff fall exactly?

RoSe: Robust Self-supervised Stereo Matching under Adverse Weather Conditions
Yun Wang, Junjie Hu, Junhui Hou, Chenghao Zhang, Renwei Yang, Dapeng Oliver Wu
https://arxiv.org/abs/2509.19165

"I actually think acting is a form of self-hypnosis. You have to be hyper, hyper aware of what's going on around you. You have to know where the lens is, what the shot is, and where you're moving. And then you have to trick yourself into an emotional state where you believe this stuff is actually happening."
—Julianne Moore
#acting

Extracting Higgs Self-Coupling Constraints through Triple Higgs Boson Production at Future Hadron Colliders
Benjamin Fuks, Andreas Papaefstathiou, Gilberto Tetlalmatzi-Xolocotzi
https://arxiv.org/abs/2509.16364

Extracting Higgs Self-Coupling Constraints through Triple Higgs Boson Production at Future Hadron Colliders
We present a systematic study of triple Higgs boson production at future high-energy hadron colliders, using the six-$b$-jet final state as a probe of the Higgs self-interactions. We conduct, under realistic detector smearing assumptions, both a traditional cut-based analysis, and a multivariate one using gradient boosting. The multivariate strategy is found to enhance sensitivity to beyond the Standard Model effects on the Higgs boson's self-couplings, while preserving large signal event yield…

3 Takeaways: Defense rescued Cowboys from self-inflicted wounds vs PHI https://cowboyswire.usatoday.com/story/sports/nfl/cowboys/2025/11/24/cowboys-eagles-3-takeaways-week-12/87442214007/

petster: Pet owner social networks (2016)
A network of friendships among users on catster.com and dogster.com. Node represents a user, and edge exists if two users self-identify as friends on either website. The dogster/catster network also contains family links between dog and dog, cat and cat, as well as dog and cat.
This network has 623766 nodes and 15699276 edges.
Tags: Social, Online, Unweighted

Alleged Scattered Spider member turns self in to Las Vegas police https://therecord.media/las-vegas-arrest-scattered-spider-suspect-turns-self-in

A Unified Formal Theory on the Logical Limits of Symbol Grounding
Zhangchi Liu
https://arxiv.org/abs/2509.20409 https://arxiv.org/pdf/2509.20409

A Unified Formal Theory on the Logical Limits of Symbol Grounding
This paper synthesizes a series of formal proofs to construct a unified theory on the logical limits of the Symbol Grounding Problem. We demonstrate through a four-stage argument that meaning within a formal system must arise from a process that is external, dynamic, and non-algorithmic. First, we prove that any purely symbolic system, devoid of external connections, cannot internally establish a consistent foundation for meaning due to self-referential paradoxes. Second, we extend this limitat…

Deblock, an app that lets users manage fiat and digital assets in one place by linking a bank account with a self-owned crypto wallet, raised a €30M Series A (Tamara Djurickovic/Tech.eu)
https://tech.eu/2025/11/19/deblock-secures-eur30m-series…

Deblock secures €30M Series A to expand on-chain banking in Europe
French crypto-banking startup Deblock has secured €30 million in Series A funding led by Speedinvest, alongside CommerzVentures and Latitude. Existing investors 20VC, Headline, Chalfen Ventures, and Kraken Ventures also participated in the round. Deblock offers a fully on-chain banking solution in Europe that combines features of a traditional bank account with on-chain self-custody. By linking a euro current account with a personal, self-owned crypto wallet, the platform allows users t…

Second-order self-force potential-region binary dynamics at $O(G^5)$ in supergravity
Zvi Bern, Enrico Herrmann, Radu Roiban, Michael S. Ruf, Alexander V. Smirnov, Vladimir A. Smirnov, Mao Zeng
https://arxiv.org/abs/2509.17412

Second-order self-force potential-region binary dynamics at $O(G^5)$ in supergravity
We compute the potential-graviton contributions to the conservative scattering angle of two non-spinning bodies in maximal supergravity at fifth order in Newton's constant, including second-order self-force effects. Our goal is to tackle the challenging integrals arising at this order in Einstein gravity, but within the technically simpler framework of supergravity. The calculation employs the scattering-amplitude framework, effective field theory, and multi-loop integration techniques based on…

Crosslisted article(s) found for cs.SE. https://arxiv.org/list/cs.SE/new
[1/1]:
- Agentic Metacognition: Designing a "Self-Aware" Low-Code Agent for Failure Prediction and Human H...
Jiexi Xu

Self-similar blowup for mass supercritical Schr\"odinger equations
Roland Donninger, Lorenz Lichtnecker
https://arxiv.org/abs/2509.16600 https://arxiv…

Self-similar blowup for mass supercritical Schrödinger equations
We consider the focusing nonlinear Schrödinger equation in three spatial dimensions with powers close to three and prove the existence of a self-similar solution. This generalizes a previous result on the cubic case and shows that self-similar blowup is stable under perturbations of the equation.

Les organisations scientifiques tirent la sonnette d'alarme sur la relation directe entre la dépendance aux écrans et l'anxiété, la dépression et, dans les cas les plus extrêmes, les tendances suicidaires.
https://en…

MACD: Multi-Agent Clinical Diagnosis with Self-Learned Knowledge for LLM
Wenliang Li, Rui Yan, Xu Zhang, Li Chen, Hongji Zhu, Jing Zhao, Junjun Li, Mengru Li, Wei Cao, Zihang Jiang, Wei Wei, Kun Zhang, Shaohua Kevin Zhou
https://arxiv.org/abs/2509.20067

MACD: Multi-Agent Clinical Diagnosis with Self-Learned Knowledge for LLM
Large language models (LLMs) have demonstrated notable potential in medical applications, yet they face substantial challenges in handling complex real-world clinical diagnoses using conventional prompting methods. Current prompt engineering and multi-agent approaches typically optimize isolated inferences, neglecting the accumulation of reusable clinical experience. To address this, this study proposes a novel Multi-Agent Clinical Diagnosis (MACD) framework, which allows LLMs to self-learn cli…

Facilitating Individuals' Sensemaking about Sedentary Behavior via Contextualized Data
Kefan Xu, Rosa I. Arriaga
https://arxiv.org/abs/2509.19420 https://

Facilitating Individuals' Sensemaking about Sedentary Behavior via Contextualized Data
The sedentary lifestyle increases individuals' risks of developing chronic diseases. To support individuals to be more physically active, we propose a mobile system, MotionShift, that presents users with step count data alongside contextual information (e.g., location, weather, calendar events, etc.) and self-reported records. By implementing and deploying this system, we aim to understand how contextual information impacts individuals' sense-making on sensor-captured data and how individuals l…

This girl now self-hosts her DNS!!
https://github.com/LuNeder/nixos-config/commit/7c5358fd4de441e4bb5011fb90f182bbbfb6bd28

The team I've cheered my whole life failed once again to make it all the way through October (but they got close!). So while my neighbors would call it fish and chips, we're having what any self-respecting Brewers fan knows is called fish fry for Game 1 - it is Friday, after all. Go Jays Go!
#baseball #WorldSeries

#ReleaseFriday — New version 3.1.0 of the recently talked about https://thi.ng/args package, a declarative & functional CLI argument parser & app framework. I updated the arg specifications to be fully self-desc…

Holographic Transformers for Complex-Valued Signal Processing: Integrating Phase Interference into Self-Attention
Enhao Huang, Zhiyu Zhang, Tianxiang Xu, Chunshu Xia, Kaichun Hu, Yuchen Yang, Tongtong Pan, Dong Dong, Zhan Qin
https://arxiv.org/abs/2509.19331

Holographic Transformers for Complex-Valued Signal Processing: Integrating Phase Interference into Self-Attention
Complex-valued signals encode both amplitude and phase, yet most deep models treat attention as real-valued correlation, overlooking interference effects. We introduce the Holographic Transformer, a physics-inspired architecture that incorporates wave interference principles into self-attention. Holographic attention modulates interactions by relative phase and coherently superimposes values, ensuring consistency between amplitude and phase. A dual-headed decoder simultaneously reconstructs the…

Sonnet 151 - CLI
Love is too young to know what conscience is,
Yet who knows not conscience is born of love?
Then, gentle cheater, urge not my amiss,
Lest guilty of my faults thy sweet self prove:
For, thou betraying me, I do betray
My nobler part to my gross body's treason;
My soul doth tell my body that he may
Triumph in love; flesh stays no farther reason,
But rising at thy name doth point out thee,
As his triumphant prize. P…

⚡ Just run https://copyparty-sfx.py - single-file self-extracting executable, no installation or setup required
🔒 HTTPS support with automatic certificate handling and reverse-proxy compatibility for nginx, Cloudflare tunnels
📝 Built-in markdown viewer and editor with syntax highlighting and live preview<…

Orbital magnetization and magnetic susceptibility of interacting electrons
Jian Kang, Minxuan Wang, Oskar Vafek
https://arxiv.org/abs/2509.20626 https://ar…

Orbital magnetization and magnetic susceptibility of interacting electrons
We present a rigorous derivation of the orbital magnetization formula for interacting electrons within the self-consistent Hartree-Fock approximation. Our results are expressed entirely in terms of the self-consistent wavefunctions and the Hartree-Fock energy spectrum at zero magnetic field. We test the formula on an interacting Rashba model, finding an agreement with calculations performed at small but non-zero external magnetic field. Our method allows us to also derive formulas for the orbit…

Analysing the prevalence of tidal features in HSC-SSP using self-supervised representation learning
A. Desmons, S. Brough, F. Lanusse, L. Canepa, A. Khalid
https://arxiv.org/abs/2509.18274

Robust self-testing of quantum steering assemblages via operator inequalities
Beata Zjawin
https://arxiv.org/abs/2509.15699 https://arxiv.org/pdf/2509.1569…

Robust self-testing of quantum steering assemblages via operator inequalities
Robust self-testing provides a framework for certifying quantum resources under experimental imperfections. Improving robustness bounds for quantum resources such as quantum states, steering assemblages, and measurements is a constant effort that ensures relevance in the experimental realm. Despite progress in analytic self-testing methods for quantum states and measurements, extending these techniques to device-independent certification of steering assemblages has remained an open challenge, w…

Pi-Transformer: A Physics-informed Attention Mechanism for Time Series Anomaly Detection
Sepehr Maleki, Negar Pourmoazemi
https://arxiv.org/abs/2509.19985 https://

Pi-Transformer: A Physics-informed Attention Mechanism for Time Series Anomaly Detection
Anomalies in multivariate time series often arise from temporal context and cross-channel coordination rather than isolated outliers. We present Pi-Transformer, a physics-informed transformer with two attention pathways: a data-driven series attention and a smoothly evolving prior attention that encodes temporal invariants such as scale-related self-similarity and phase synchrony. The prior acts as a stable reference that calibrates reconstruction error. During training, we pair a reconstructio…

Agentic Metacognition: Designing a "Self-Aware" Low-Code Agent for Failure Prediction and Human Handoff
Jiexi Xu
https://arxiv.org/abs/2509.19783 https://

Agentic Metacognition: Designing a "Self-Aware" Low-Code Agent for Failure Prediction and Human Handoff
The inherent non-deterministic nature of autonomous agents, particularly within low-code/no-code (LCNC) environments, presents significant reliability challenges. Agents can become trapped in unforeseen loops, generate inaccurate outputs, or encounter unrecoverable failures, leading to user frustration and a breakdown of trust. This report proposes a novel architectural pattern to address these issues: the integration of a secondary, "metacognitive" layer that actively monitors the primary LCNC…

Integrated Framework for LLM Evaluation with Answer Generation
Sujeong Lee, Hayoung Lee, Seongsoo Heo, Wonik Choi
https://arxiv.org/abs/2509.20097 https://…

Integrated Framework for LLM Evaluation with Answer Generation
Reliable evaluation of large language models is essential to ensure their applicability in practical scenarios. Traditional benchmark-based evaluation methods often rely on fixed reference answers, limiting their ability to capture important qualitative aspects of generated responses. To address these shortcomings, we propose an integrated evaluation framework called \textit{self-refining descriptive evaluation with expert-driven diagnostics}, SPEED, which utilizes specialized functional expert…

Chinese state media: China's new five-year development plan aims to achieve tech self-reliance and expand the domestic market, after a four-day CCP conclave (Bloomberg)
https://www.bloomberg.com/news/articles/2025-10-23/…

Self-Alignment Learning to Improve Myocardial Infarction Detection from Single-Lead ECG
Jiarui Jin, Xiaocheng Fang, Haoyu Wang, Jun Li, Che Liu, Donglin Xie, Hongyan Li, Shenda Hong
https://arxiv.org/abs/2509.19397

Self-Alignment Learning to Improve Myocardial Infarction Detection from Single-Lead ECG
Myocardial infarction is a critical manifestation of coronary artery disease, yet detecting it from single-lead electrocardiogram (ECG) remains challenging due to limited spatial information. An intuitive idea is to convert single-lead into multiple-lead ECG for classification by pre-trained models, but generative methods optimized at the signal level in most cases leave a large latent space gap, ultimately degrading diagnostic performance. This naturally raises the question of whether latent s…

Linking Young Stellar Object Morphology to Evolutionary Stages with Self-Organizing Maps
David Hernandez, Odysseas Dionatos, Marc Audard, G\'abor Marton, Julia Roquette, Ilknur Gezer, M\'at\'e Madar\'asz, Kai L. Polsterer
https://arxiv.org/abs/2509.19069

Linking Young Stellar Object Morphology to Evolutionary Stages with Self-Organizing Maps
Studies in the past few decades have investigated young stellar object evolution based on their spectral energy distribution (SED). The SED is heavily influenced not only by evolutionary stage, but also the morphology of the young star. This work is part of the NEMESIS project which is aiming to revisit star formation with the aid of machine learning techniques and provides the framework for this work. In a first effort towards a novel spectro-morphological classification we analyzed young stel…

It's #MusicWomenWednesday again, and the self titled debut EP by Spike (the synthpop project of UK artist Hannah McLoughlin) is a treat. The first track is a wild, synthy cover of Warren Zevon's 'Werewolves of London', and the other tracks are excellent catchy earworms full of cool hooks. An EP that goes by too fast. Would love an LP of this.

Agentic Metacognition: Designing a "Self-Aware" Low-Code Agent for Failure Prediction and Human Handoff
Jiexi Xu
https://arxiv.org/abs/2509.19783 https://

Agentic Metacognition: Designing a "Self-Aware" Low-Code Agent for Failure Prediction and Human Handoff
The inherent non-deterministic nature of autonomous agents, particularly within low-code/no-code (LCNC) environments, presents significant reliability challenges. Agents can become trapped in unforeseen loops, generate inaccurate outputs, or encounter unrecoverable failures, leading to user frustration and a breakdown of trust. This report proposes a novel architectural pattern to address these issues: the integration of a secondary, "metacognitive" layer that actively monitors the primary LCNC…

Non-renormalisation of coupling constants from categorical symmetries in two dimensions
Guillermo Arias-Tamargo, Chris Hull, Maxwell L. Vel\'asquez Cotini Hutt
https://arxiv.org/abs/2509.20441

Non-renormalisation of coupling constants from categorical symmetries in two dimensions
We study the role of categorical symmetries in constraining the renormalisation of couplings in two-dimensional non-linear sigma models with Wess-Zumino term. A large class of these theories admit self-duality symmetries associated with discrete gauging and T-duality. They are generically non-conformal, but we argue that a particular coupling is protected from quantum corrections by the categorical symmetry. We give strong evidence for this claim by showing that the $β$-function for this coupl…

Enhancing Semantic Segmentation with Continual Self-Supervised Pre-training
Brown Ebouky, Ajad Chhatkuli, Cristiano Malossi, Christoph Studer, Roy Assaf, Andrea Bartezzaghi
https://arxiv.org/abs/2509.17816

Enhancing Semantic Segmentation with Continual Self-Supervised Pre-training
Self-supervised learning (SSL) has emerged as a central paradigm for training foundation models by leveraging large-scale unlabeled datasets, often producing representations with strong generalization capabilities. These models are typically pre-trained on general-purpose datasets such as ImageNet and subsequently adapted to various downstream tasks through finetuning. While recent advances have explored parameter-efficient strategies for adapting pre-trained models, extending SSL pre-training …

Sonnet 087 - LXXXVII
Farewell! thou art too dear for my possessing,
And like enough thou know'st thy estimate,
The charter of thy worth gives thee releasing;
My bonds in thee are all determinate.
For how do I hold thee but by thy granting?
And for that riches where is my deserving?
The cause of this fair gift in me is wanting,
And so my patent back again is swerving.
Thy self thou gavest, thy own worth then not knowing,
Or me to…

Quantum statistical mechanical gauge invariance
Johanna M\"uller, Matthias Schmidt
https://arxiv.org/abs/2509.20494 https://arxiv.org/pdf/2509.20494…

Quantum statistical mechanical gauge invariance
We address gauge invariance in the statistical mechanics of quantum many-body systems. The gauge transformation acts on the position and momentum degrees of freedom and it is represented by a quantum shifting superoperator that maps quantum observables onto each other. The shifting superoperator is anti-self-adjoint and it has noncommutative Lie algebra structure. These properties induce exact equilibrium sum rules that connect locally-resolved force and hyperforce densities for any given obser…

Closed Walks Of Low Dimension And Twisted Moments On Self-Loop Graphs
Johnny Lim
https://arxiv.org/abs/2509.17035 https://arxiv.org/pdf/2509.17035

Closed Walks Of Low Dimension And Twisted Moments On Self-Loop Graphs
Let $G_S$ be a graph with loops attached at each vertex in $S \subseteq V(G).$ In this article, we develop exact formulae for the number of closed $3$- and $4$-walks on $G_S$ in terms of vertex degrees and certain elementary subgraphs of $G_S.$ We then derive the specific closed walks formulae for several graph families such as complete bipartite self-loop graphs, complete graphs, cycle graphs, etc. We demonstrate that such invariants are non-trivial in $G_S,$ which otherwise may be trivial in …

Calibrated Reasoning: An Explanatory Verifier for Dynamic and Efficient Problem-Solving
Anisha Garg, Engin Tekin, Yash More, David Bick, Nishit Neema, Ganesh Venkatesh
https://arxiv.org/abs/2509.19681 …

Calibrated Reasoning: An Explanatory Verifier for Dynamic and Efficient Problem-Solving
Advanced test-time computing strategies are essential for scaling reasoning models, but their effectiveness is capped by the models' poor self-evaluation. We propose a pairwise Explanatory Verifier, trained via reinforcement learning (GRPO), that produces calibrated confidence scores and associated natural language reasoning for generated solutions. Our verifier improves the accuracy and efficiency of test-time strategies like best-of-n and self-reflection. Crucially, it excels at identifying c…

Blind-Spot Guided Diffusion for Self-supervised Real-World Denoising
Shen Cheng, Haipeng Li, Haibin Huang, Xiaohong Liu, Shuaicheng Liu
https://arxiv.org/abs/2509.16091 https://…

Blind-Spot Guided Diffusion for Self-supervised Real-World Denoising
In this work, we present Blind-Spot Guided Diffusion, a novel self-supervised framework for real-world image denoising. Our approach addresses two major challenges: the limitations of blind-spot networks (BSNs), which often sacrifice local detail and introduce pixel discontinuities due to spatial independence assumptions, and the difficulty of adapting diffusion models to self-supervised denoising. We propose a dual-branch diffusion framework that combines a BSN-based diffusion branch, generati…

Calibrated Reasoning: An Explanatory Verifier for Dynamic and Efficient Problem-Solving
Anisha Garg, Engin Tekin, Yash More, David Bick, Nishit Neema, Ganesh Venkatesh
https://arxiv.org/abs/2509.19681 …

Calibrated Reasoning: An Explanatory Verifier for Dynamic and Efficient Problem-Solving
Advanced test-time computing strategies are essential for scaling reasoning models, but their effectiveness is capped by the models' poor self-evaluation. We propose a pairwise Explanatory Verifier, trained via reinforcement learning (GRPO), that produces calibrated confidence scores and associated natural language reasoning for generated solutions. Our verifier improves the accuracy and efficiency of test-time strategies like best-of-n and self-reflection. Crucially, it excels at identifying c…

Efficient Cell Painting Image Representation Learning via Cross-Well Aligned Masked Siamese Network
Pin-Jui Huang, Yu-Hsuan Liao, SooHeon Kim, NoSeong Park, JongBae Park, DongMyung Shin
https://arxiv.org/abs/2509.19896

Efficient Cell Painting Image Representation Learning via Cross-Well Aligned Masked Siamese Network
Computational models that predict cellular phenotypic responses to chemical and genetic perturbations can accelerate drug discovery by prioritizing therapeutic hypotheses and reducing costly wet-lab iteration. However, extracting biologically meaningful and batch-robust cell painting representations remains challenging. Conventional self-supervised and contrastive learning approaches often require a large-scale model and/or a huge amount of carefully curated data, still struggling with batch ef…

Self-Supervised Cross-Modal Learning for Image-to-Point Cloud Registration
Xingmei Wang, Xiaoyu Hu, Chengkai Huang, Ziyan Zeng, Guohao Nie, Quan Z. Sheng, Lina Yao
https://arxiv.org/abs/2509.15882

Self-Supervised Cross-Modal Learning for Image-to-Point Cloud Registration
Bridging 2D and 3D sensor modalities is critical for robust perception in autonomous systems. However, image-to-point cloud (I2P) registration remains challenging due to the semantic-geometric gap between texture-rich but depth-ambiguous images and sparse yet metrically precise point clouds, as well as the tendency of existing methods to converge to local optima. To overcome these limitations, we introduce CrossI2P, a self-supervised framework that unifies cross-modal learning and two-stage reg…

Towards Sharper Object Boundaries in Self-Supervised Depth Estimation
Aur\'elien Cecille, Stefan Duffner, Franck Davoine, R\'emi Agier, Thibault Neveu
https://arxiv.org/abs/2509.15987

Towards Sharper Object Boundaries in Self-Supervised Depth Estimation
Accurate monocular depth estimation is crucial for 3D scene understanding, but existing methods often blur depth at object boundaries, introducing spurious intermediate 3D points. While achieving sharp edges usually requires very fine-grained supervision, our method produces crisp depth discontinuities using only self-supervision. Specifically, we model per-pixel depth as a mixture distribution, capturing multiple plausible depths and shifting uncertainty from direct regression to the mixture w…

UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation
Xiaoqi Zhao, Youwei Pang, Chenyang Yu, Lihe Zhang, Huchuan Lu, Shijian Lu, Georges El Fakhri, Xiaofeng Liu
https://arxiv.org/abs/2509.16170

UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation
Multi-modal image segmentation faces real-world deployment challenges from incomplete/corrupted modalities degrading performance. While existing methods address training-inference modality gaps via specialized per-combination models, they introduce high deployment costs by requiring exhaustive model subsets and model-modality matching. In this work, we propose a unified modality-relax segmentation network (UniMRSeg) through hierarchical self-supervised compensation (HSSC). Our approach hierarch…