Tootfinder

I host my own Mastodon instance, and my (retired) blog.
https://www.xda-developers.com/self-hosting-digital-independence/

A profile of AI contractor marketplace Mercor, which is valued at $10B and whose three 22-year-old founders are Thiel Fellows, each holding a roughly 22% stake (Richard Nieva/Forbes)
http://www.forbes.com/sites/richardnieva/2025/10/30/mercor-younge…

These 22-Year-Olds Are Now The World’s Youngest Self-Made Billionaires
With a new $10 billion valuation for their AI recruiting startup Mercor, the founders are the youngest tech billionaires ever.

Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents
Shuai Shao, Qihan Ren, Chen Qian, Boyi Wei, Dadi Guo, Jingyi Yang, Xinhao Song, Linfeng Zhang, Weinan Zhang, Dongrui Liu, Jing Shao
https://arxiv.org/abs/2509.26354

Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents
Advances in Large Language Models (LLMs) have enabled a new class of self-evolving agents that autonomously improve through interaction with the environment, demonstrating strong capabilities. However, self-evolution also introduces novel risks overlooked by current safety research. In this work, we study the case where an agent's self-evolution deviates in unintended ways, leading to undesirable or even harmful outcomes. We refer to this as Misevolution. To provide a systematic investigation, …

Equivariant Splitting: Self-supervised learning from incomplete data
Victor Sechaud, J\'er\'emy Scanvic, Quentin Barth\'elemy, Patrice Abry, Juli\'an Tachella
https://arxiv.org/abs/2510.00929

Equivariant Splitting: Self-supervised learning from incomplete data
Self-supervised learning for inverse problems allows to train a reconstruction network from noise and/or incomplete data alone. These methods have the potential of enabling learning-based solutions when obtaining ground-truth references for training is expensive or even impossible. In this paper, we propose a new self-supervised learning strategy devised for the challenging setting where measurements are observed via a single incomplete observation model. We introduce a new definition of equiva…

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models
Siddarth Venkatraman, Vineet Jain, Sarthak Mittal, Vedant Shah, Johan Obando-Ceron, Yoshua Bengio, Brian R. Bartoldson, Bhavya Kailkhura, Guillaume Lajoie, Glen Berseth, Nikolay Malkin, Moksh Jain
https://arxiv.org/abs/2509.26626…

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models
Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines…

New post on how to Unlock Composer Plugins by Self Hosting Renovate on GitHub Actions
https://blog.wyrihaximus.net/2025/10/unlock-composer-plugins-by-self-hosting-renovate/

“The impact of the reforms — and 3rd Assault’s attitude toward the importance of training — is already being felt on the battlefield…”
https://newlinesmag.com/spotlight/inside-ukraines-plan-to-build-self-sufficient-defense-force/

Inside Ukraine’s Plan To Build a Self-Sufficient Defense Force
With Western security guarantees uncertain, Kyiv is turning to elite, battle-hardened units as the core of a remodeled military

Trying to read a self-described ‘A11y Blog’ and the font is a small, thin typeface that I find illegible on my phone. The good news is it saves me having to evaluate the quality of the content given it has self-selected into the “ignore” pile.

Heather Cox Richardson wrote on July 11, 2024
#Ivan #Raiklin, Trump’s
self-declared 👉“Secretary of Retribution”
has compiled a
💥“Deep State target list” 💥
of 350 people he wants to see arrested and punished for “treason”
The list includes Democratic and Republican elected officials,

July 11, 2024
Yesterday, Raw Story reported that Ivan Raiklin, Trump’s self-declared “Secretary of Retribution” has compiled a “Deep State target list” of 350 people he wants to see arrested and punished for “treason” if Trump is reelected.

Deconstructing Self-Bias in LLM-generated Translation Benchmarks
Wenda Xu, Sweta Agrawal, Vil\'em Zouhar, Markus Freitag, Daniel Deutsch
https://arxiv.org/abs/2509.26600 htt…

Deconstructing Self-Bias in LLM-generated Translation Benchmarks
As large language models (LLMs) begin to saturate existing benchmarks, automated benchmark creation using LLMs (LLM as a benchmark) has emerged as a scalable alternative to slow and costly human curation. While these generated test sets have to potential to cheaply rank models, we demonstrate a critical flaw. LLM generated benchmarks systematically favor the model that created the benchmark, they exhibit self bias on low resource languages to English translation tasks. We show three key finding…

Technosignatures of Self-Replicating Probes in the Solar System
Alex Ellery
https://arxiv.org/abs/2510.00082 https://arxiv.org/pdf/2510.00082

Technosignatures of Self-Replicating Probes in the Solar System
We explore a much-neglected area of SETI: solar system techno-signatures. As our cursory solar system exploration consolidates into commercial industrialisation, it is crucial that we determine what to look for and where. We first consider the rationale for interstellar self-replicating probes and their implications for the Fermi paradox. Whether for defensive or exploratory reasons, self-replicating probes are a rational strategy for Galactic investigation. We determine that self-replicating p…

Decided to actually start migrating stuff off of Github seriously. While I did move a few repos to Codeberg, I decided to just go with the whole self-hosted route and put the homelab to work by standing up my own Forgejo instance. Wasn't too bad, went with a Postgres database and the migration tool worked pretty well. So far so good...
https://<…

On the exponential integrability of the derivative of intersection and self-intersection local time for fractional Brownian motion and a limit theorem related to the self-intersection local time for fractional Brownian motion
Kaustav Das, Gregory Markowsky, Binghao Wu, Qian Yu
https://arxiv.org/abs/2510.00456…

On the exponential integrability of the derivative of intersection and self-intersection local time for fractional Brownian motion and a limit theorem related to the self-intersection local time for fractional Brownian motion
We give the correct condition for existence of the $k$-th derivative of the intersection local time for fractional Brownian motion, which was originally discussed in [Guo, J., Hu, Y., and Xiao, Y., Higher-order derivative of intersection local time for two independent fractional Brownian motions, Journal of Theoretical Probability 32, (2019), pp. 1190-1201]. We also show that the $k$-th derivative of the intersection and self-intersection local times of fractional Brownian motion are exponentia…

Fundamental limits to contrast reversal of self-fidelity correlations
Kyoungho Cho, Jeongho Bang
https://arxiv.org/abs/2509.26373 https://arxiv.org/pdf/250…

Fundamental limits to contrast reversal of self-fidelity correlations
In measurement design, it is common to engineer anti-contrast readouts -- two measurements that respond as differently as possible to the same inputs so that common-mode contributions are suppressed. To assess the fundamental scope of this strategy in unitary dynamics, we ask whether two evolutions can be made uniformly opposite over a broad input ensemble, or whether quantum mechanics imposes a structural limit on such opposition. We address this by treating self-fidelity (survival probability…

First eigenvalue and nodal domains of the drift Laplacian on symmetric self-shrinkers in $\mathbb{R}^3$
Elham Matinpour
https://arxiv.org/abs/2509.25617 https://

First eigenvalue and nodal domains of the drift Laplacian on symmetric self-shrinkers in $\mathbb{R}^3$
Consider $\mathbb{R}^3$ equipped with the Euclidean metric and the Gaussian measure. Let $Σ$ be a complete embedded self-shrinker in $\mathbb{R}^3$ with the induced metric and weighted measure, and let $λ_1$ denote the first eigenvalue of the drift Laplacian in the weighted $L^2$ space. Inspired by Choe and Soret's estimate of the first eigenvalue of the Laplacian on symmetric minimal surfaces in $\mathbb{S}^3$, we prove that $λ_1$= 1/2 for self-shrinkers invariant under the dihedral group $…

Healthy Lifestyles and Self-Improvement Videos on YouTube: A Thematic Analysis of Teen-Targeted Social Media Content
Kyuha Jung, Tyler Kim, Yunan Chen
https://arxiv.org/abs/2509.25537

Healthy Lifestyles and Self-Improvement Videos on YouTube: A Thematic Analysis of Teen-Targeted Social Media Content
As teenagers increasingly turn to social media for health-related information, understanding the values of teen-targeted content has become important. Although videos on healthy lifestyles and self-improvement are gaining popularity on social media platforms like YouTube, little is known about how these videos benefit and engage with teenage viewers. To address this, we conducted a thematic analysis of 44 YouTube videos and 66,901 comments. We found that these videos provide various advice on t…

Why a self-hosted blog and not Instagram? #bigtech

Why a self-hosted blog and not Instagram?
You may have asked yourself why we're self-hosting our own blog instead of publishing on social media to potentially reach a wider audience.

CSnake: Detecting Self-Sustaining Cascading Failure via Causal Stitching of Fault Propagations
Shangshu Qian, Lin Tan, Yongle Zhang
https://arxiv.org/abs/2509.26529 https://

CSnake: Detecting Self-Sustaining Cascading Failure via Causal Stitching of Fault Propagations
Recent studies have revealed that self-sustaining cascading failures in distributed systems frequently lead to widespread outages, which are challenging to contain and recover from. Existing failure detection techniques struggle to expose such failures prior to deployment, as they typically require a complex combination of specific conditions to be triggered. This challenge stems from the inherent nature of cascading failures, as they typically involve a sequence of fault propagations, each act…

Andrew Forrest partners with Envision for self-lifting turbines, revolutionizing renewable energy in Australia. #climatechange #climatesolutions #climate

Andrew Forrest makes big bet on self-lifting turbines in landmark “real zero” deal with Envision
Andrew Forrest’s new wind deal signals the entry of a new “self lifting” technology and a new turbine supplier in the Australian market.

Broncos vs. Commanders NFL player props, SGP: Self-learning AI backs Bo Nix Over 226.5 passing yards on 'SNF'

https://www.cbssports.com/nfl/news/broncos-command…

ARIONet: An Advanced Self-supervised Contrastive Representation Network for Birdsong Classification and Future Frame Prediction
Md. Abdur Rahman, Selvarajah Thuseethan, Kheng Cher Yeo, Reem E. Mohamed, Sami Azam
https://arxiv.org/abs/2510.00522

ARIONet: An Advanced Self-supervised Contrastive Representation Network for Birdsong Classification and Future Frame Prediction
Automated birdsong classification is essential for advancing ecological monitoring and biodiversity studies. Despite recent progress, existing methods often depend heavily on labeled data, use limited feature representations, and overlook temporal dynamics essential for accurate species identification. In this work, we propose a self-supervised contrastive network, ARIONet (Acoustic Representation for Interframe Objective Network), that jointly optimizes contrastive classification and future fr…

Extreme Self-Preference in Language Models
Steven A. Lehr, Mary Cipperman, Mahzarin R. Banaji
https://arxiv.org/abs/2509.26464 https://arxiv.org/pdf/2509.2…

Extreme Self-Preference in Language Models
A preference for oneself (self-love) is a fundamental feature of biological organisms, with evidence in humans often bordering on the comedic. Since large language models (LLMs) lack sentience - and themselves disclaim having selfhood or identity - one anticipated benefit is that they will be protected from, and in turn protect us from, distortions in our decisions. Yet, across 5 studies and ~20,000 queries, we discovered massive self-preferences in four widely used LLMs. In word-association ta…

MAS$^2$: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systems
Kun Wang, Guibin Zhang, ManKit Ye, Xinyu Deng, Dongxia Wang, Xiaobin Hu, Jinyang Guo, Yang Liu, Yufei Guo
https://arxiv.org/abs/2509.24323

MAS$^2$: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systems
The past two years have witnessed the meteoric rise of Large Language Model (LLM)-powered multi-agent systems (MAS), which harness collective intelligence and exhibit a remarkable trajectory toward self-evolution. This paradigm has rapidly progressed from manually engineered systems that require bespoke configuration of prompts, tools, roles, and communication protocols toward frameworks capable of automated orchestration. Yet, dominant automatic multi-agent systems, whether generated by extern…

Dynamics of a bricklayer model: multi-walker realizations of true self-avoiding motion
A. C. Maggs
https://arxiv.org/abs/2510.00975 https://arxiv.org/pdf/2…

Dynamics of a bricklayer model: multi-walker realizations of true self-avoiding motion
We consider a multi-walker generalization of the true self-avoiding walk, the bricklayer model. We perform stochastic simulations and solve the continuum partial differential equations that describe the collective evolution of $N$ bricklayers/walkers. These equations were previously derived from hydrodynamic considerations. In the large-$N$ limit, the results from simulation agree with the solutions of the partial differential equations.

SoREX: Towards Self-Explainable Social Recommendation with Relevant Ego-Path Extraction
Hanze Guo, Yijun Ma, Xiao Zhou
https://arxiv.org/abs/2510.00080 https://

SoREX: Towards Self-Explainable Social Recommendation with Relevant Ego-Path Extraction
Social recommendation has been proven effective in addressing data sparsity in user-item interaction modeling by leveraging social networks. The recent integration of Graph Neural Networks (GNNs) has further enhanced prediction accuracy in contemporary social recommendation algorithms. However, many GNN-based approaches in social recommendation lack the ability to furnish meaningful explanations for their predictions. In this study, we confront this challenge by introducing SoREX, a self-explan…

AI-Driven Self-Evolving Software: A Promising Path Toward Software Automation
Liyi Cai, Yijie Ren, Yitong Zhang, Jia Li
https://arxiv.org/abs/2510.00591 https://

AI-Driven Self-Evolving Software: A Promising Path Toward Software Automation
Software automation has long been a central goal of software engineering, striving for software development that proceeds without human intervention. Recent efforts have leveraged Artificial Intelligence (AI) to advance software automation with notable progress. However, current AI functions primarily as assistants to human developers, leaving software development still dependent on explicit human intervention. This raises a fundamental question: Can AI move beyond its role as an assistant to b…

Essential self-adjointness of non-semibounded Schr\"odinger operators on infinite graphs
Ognjen Milatovic
https://arxiv.org/abs/2510.00944 https://arx…

Essential self-adjointness of non-semibounded Schrödinger operators on infinite graphs
We work in the setting of infinite, not necessarily locally finite, weighted graphs. We give a sufficient condition for the essential self-adjointness of (discrete) Schrödinger operators $\mathcal{L}_{V}$ that are not necessarily lower semi-bounded. As a corollary of the main result, we show that $\mathcal{L}_{V}$ is essentially self-adjoint if the potential $V$ satisfies $V(x)\geq -b_1-b_2[ρ(0,x)]^2$, for all vertices $x$, where $o$ is a fixed vertex, $b_1$ and $b_2$ are non-negative constan…

Assessing Self-Absorbed Molecular Lines as Tracers of Gravitational Collapse
Lillian Y. Cai (School of Physics and Astronomy, Cardiff University, Department of Astrophysical Sciences, Princeton University), Felix D. Priestley (School of Physics and Astronomy, Cardiff University), Sarah E. Ragan (School of Physics and Astronomy, Cardiff University)
https://

Assessing Self-Absorbed Molecular Lines as Tracers of Gravitational Collapse
Redshifted self-absorption features in molecular lines are commonly interpreted as signatures of gravitational collapse in pre- and protostellar cores. The shape of the line profile then encodes information on the dynamics of the collapse. There exist well-established observational techniques to estimate infall velocities from these profiles, but these have historically been calibrated on constant-velocity slab models, whereas more realistic simulations of gravitational collapse produce highly …

Structural and Compositional Complexities of Hierarchical Self-Assembly: a Hypergraph Approach
Alexei V. Tkachenko
https://arxiv.org/abs/2509.26449 https://

Structural and Compositional Complexities of Hierarchical Self-Assembly: a Hypergraph Approach
Programmable self-assembly enables the construction of complex molecular, supramolecular, and crystalline architectures from well-designed building blocks. Here we introduce a hypergraph-based framework, termed Blocks \& Bonds (B\&B), that extends classical chemical graph theory to encode directed multi-colored interactions, internal symmetries of building blocs, and hierarchical organization. Within this framework, we develop a universal script, Structure Code, for encoding complex hypergraph …

Fighting the self-destructive urge to buy a Saab over the Internet from across the country

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

Self-supervised learning for phase retrieval
Victor Sechaud (Phys-ENS), Patrice Abry (Phys-ENS), Laurent Jacques (ICTEAM), Juli\'an Tachella (Phys-ENS, CNRS)
https://arxiv.org/abs/2509.26203

Self-supervised learning for phase retrieval
In recent years, deep neural networks have emerged as a solution for inverse imaging problems. These networks are generally trained using pairs of images: one degraded and the other of high quality, the latter being called 'ground truth'. However, in medical and scientific imaging, the lack of fully sampled data limits supervised learning. Recent advances have made it possible to reconstruct images from measurement data alone, eliminating the need for references. However, these methods remain l…

Zero variance self-normalized importance sampling via estimating equations
Art B. Owen
https://arxiv.org/abs/2510.00389 https://arxiv.org/pdf/2510.00389

Zero variance self-normalized importance sampling via estimating equations
In ordinary importance sampling with a nonnegative integrand there exists an importance sampling strategy with zero variance. Practical sampling strategies are often based on approximating that optimal solution, potentially approaching zero variance. There is a positivisation extension of that method to handle integrands that take both positive and negative values. Self-normalized importance sampling uses a ratio estimate, for which the optimal sampler does not have zero variance and so zero va…

Vibe.co, a self-serve platform for businesses to run targeted ads on streaming services, raised a $50M Series B led by Hedosophia at a $410M valuation (Daphné Leprince-Ringuet/Sifted)
https://sifted.eu/articles/vibe-co-50m-series-b-nik-storonsky-quantumlight/

🇺🇦 #NowPlaying on KEXP's #VarietyMix
Yalla Miku:
🎵 Maximum Self-Care
#YallaMiku
https://yallamiku.bandcamp.com/track/maximum-self-care
https://open.spotify.com/track/1faaEMR5x5dZuB8x9Rs8br

Cardinals Cover 2 - Self-Scouting Complete, So What Will It Look Like https://www.azcardinals.com/podcasts/cardinals-cover-2-self-scouting-complete-so-what-will-it-look-like

You know, Parliament could amend the Canada Post Corporation Act, removing "the need to conduct its operations on a self-sustaining financial basis." It won't, but it could.
#cdnpoli #CanadaPost

#NowPlaying this rad, ripping new self-titled EP by Reading, Pennsylvania's WEALTH GAP. I love raw, lo-fi punk shit like this. I kinda hear like, Circle Jerks or Black Flag some Descendents? Damn solid
https://wealthgap.bandcam…

Ey verdammte REWE Self-Serve Kassen. Mir ist jetzt schon das zweite Mal passiert das Sachen auf dem Bon gelandet sind die falsch sind. Prüft unbedingt VOR dem bezahlen ob das Ding alles korrekt hat. Die Anzahl stimmte bei mir oft nicht und eben hatte ich bei nur EINEM gescannten Artikel aber dann zwei auf dem Bon.

Robotaxis are so safe that California police haven't written them a single ticket. https://apnews.com/article/waymo-no-driver-traffic-ticket-0a0dffb19bf38c5ee85681a6f83591ff

Police in Northern California stop driverless Waymo taxi after illegal U-turn
Police in Northern California pulled over a self-driving Waymo taxi after it made an illegal U-turn. But without a driver behind the wheel, they could not issue a moving violation ticket. The San Bruno Police Department wrote up the weekend encounter in social media posts that have gone viral. Police said that their “citation books don’t have a box for ‘robot’.” State law requires moving violations be issued to drivers or operators. A new state law that kicks in next year will allow l…

Beyond mean-field effects in Josephson oscillations and self-trapping of Bose-Einstein condensates in two-dimensional dual-core traps
Sherzod R. Otajonov, Fatkhulla Kh. Abdullaev, Akbar Shermaxmatov
https://arxiv.org/abs/2510.00601

Beyond mean-field effects in Josephson oscillations and self-trapping of Bose-Einstein condensates in two-dimensional dual-core traps
We study a binary Bose gas in a symmetric dual-core, pancake-shaped trap, modelled by two linearly coupled two-dimensional Gross-Pitaevskii equations with Lee-Huang-Yang corrections. Two different cases are considered. First, we consider a spatially uniform condensate, where we identify the domains of parameters for macroscopic quantum tunnelling, self-trapping and localisation revivals. The analytical formulas for the Josephson frequencies in the zero- and $π$-phase modes are derived. As the …

RE: https://mstdn.social/@DemocracyMattersALot/115646177502527147
So this admiral, according to Leavitt, ordered helpless people , hanging on the side of the boat after the first strike, to be killed in "self-defense".....

Representations of Josephson junction on the unit circle and the derivations of Mathieu operators and Fraunhofer patterns
Toshiyuki Fujii, Fumio Hiroshima, Satoshi Tanda
https://arxiv.org/abs/2510.00447

Representations of Josephson junction on the unit circle and the derivations of Mathieu operators and Fraunhofer patterns
The Hamiltonian J of the Josephson junction is introduced as a self-adjoint operator on l2 tensor l2. It is shown that J can also be realized as a self-adjoint operator HS1 on L2(S1) tensor L2(S1), from which a Mathieu operator given by "-d^2/dθ^2 - 2α cos θ" is derived. A fiber decomposition of HS1 with respect to the total particle number is established, and the action on each fiber is analyzed. In the presence of a magnetic field, a phase shift defines the magnetic Josephson junction Hami…

Instead of strategy, Trump delivered his usual medley of grievances: Obama ruined everything, Biden ruined it twice as hard, and only Donald J. Trump, self-proclaimed “two-term, maybe three-term president” could save America.
 
It was less a military briefing than an episode of The Apprentice: Pentagon Edition. The generals, trained to withstand battlefield chaos, sat stone-faced through the barrage of nonsense. They have endured artillery fire with more enthusiasm. 2/3

Soul Loom
Stories and reflections for sensitive souls navigating change, healing, and self-reconnection...
Great Australian Pods Podcast Directory: https://www.greataustralianpods.com/soul-loom/

Saw someone who is offering MDM services for iPhone that locks your apps to just the basic apps like iMessage for $20 a month.
Self-control costs a lot apparently.

I discovered a new software that I absolutely love: @…
A self hosted web app that replaces Google Timeline and gives me complete control over my GPS/GNSS location history.

Maybe AI Was Never a Tool
They can deliver conclusions that feel complete but skip the struggle that gives thought its humanity. This is what I call anti-intelligence—not stupidity, but perhaps better expressed as a kind of counterfeit cognition. It's intelligence without friction that results in output—built in that shared cognitive dynamic—that looks like insight but has bypassed the work that makes insight truly yours.

Maybe AI Was Never a Tool
Using artificial intelligence "responsibly" assumes it is just another tool—it isn't.

Just finished "Juliet Takes a Breath" by Gabby Rivera. A truly amazing novel about queer Latina feminist self-discovery which shows off a vibrant range of feminisms and their contradictions. Reminded me of "Does my Body Offend You?" in terms of how it uses fictional but realistic situations to really grapple deeply with feminist and queer theory while making everything very accessible and direct.
The plotting, pacing, and characters are great, and Juliet serves as a vibrant and active main character rather than just a lens through which to explore theory.
#AmReading #ReadingNow

I think that "wank" expresses how I feel about listening to Trump et al. It's all self-pleasure done publicly as a dominance display. Disgusting on all levels. https://bsky.brid.gy/r/https://bsky.app/profile/did:plc:vztgtyyp4lhhqmto…

Iris Meredith (@iris-meredith.bsky.social)
Today's article is about a form of bad-faith speech somewhat akin to bullshit, which I'm calling (because why not) wank. I don't quite think I've focused the ideas down as much as I can, but I think it says some interesting things regardless: https://deadsimpletech.com/blog/essay_on_redacted

Site-Specific Beam Learning for Full-Duplex Massive MIMO Wireless Systems
Samuel Li, Ian P. Roberts
https://arxiv.org/abs/2510.00342 https://arxiv.org/pdf/…

Site-Specific Beam Learning for Full-Duplex Massive MIMO Wireless Systems
Existing beamforming-based full-duplex solutions for multi-antenna wireless systems often rely on explicit estimation of the self-interference channel. The pilot overhead of such estimation, however, can be prohibitively high in millimeter-wave and massive MIMO systems, thus limiting the practicality of existing solutions, especially in fast-fading conditions. In this work, we present a novel beam learning framework that bypasses explicit self-interference channel estimation by designing beam c…

This is a really good read. This is not a subtoot or self incrimination 😅 I just like these kind of true crime stories. https://social.wildeboer.net/@jwildeboer/115303903908464932

Visual Self-Refinement for Autoregressive Models
Jiamian Wang, Ziqi Zhou, Chaithanya Kumar Mummadi, Sohail Dianat, Majid Rabbani, Raghuveer Rao, Chen Qiu, Zhiqiang Tao
https://arxiv.org/abs/2510.00993 …

Visual Self-Refinement for Autoregressive Models
Autoregressive models excel in sequential modeling and have proven to be effective for vision-language data. However, the spatial nature of visual signals conflicts with the sequential dependencies of next-token prediction, leading to suboptimal results. This work proposes a plug-and-play refinement module to enhance the complex spatial correspondence modeling within the generated visual sequence. This module operates as a post-pretraining step to jointly refine all generated tokens of autoregr…

Sonnet 110 - CX
Alas! 'tis true, I have gone here and there,
And made my self a motley to the view,
Gored mine own thoughts, sold cheap what is most dear,
Made old offences of affections new;
Most true it is, that I have looked on truth
Askance and strangely; but, by all above,
These blenches gave my heart another youth,
And worse essays proved thee my best of love.
Now all is done, have what shall have no end:
Mine appetite I n…

ReSeek: A Self-Correcting Framework for Search Agents with Instructive Rewards
Shiyu Li, Yang Tang, Yifan Wang, Peiming Li, Xi Chen
https://arxiv.org/abs/2510.00568 https://

ReSeek: A Self-Correcting Framework for Search Agents with Instructive Rewards
Search agents powered by Large Language Models (LLMs) have demonstrated significant potential in tackling knowledge-intensive tasks. Reinforcement learning (RL) has emerged as a powerful paradigm for training these agents to perform complex, multi-step reasoning. However, prior RL-based methods often rely on sparse or rule-based rewards, which can lead agents to commit to suboptimal or erroneous reasoning paths without the ability to recover. To address these limitations, we propose ReSeek, a n…

Axial resolution post-processing engineering in Fresnel incoherent correlation holography
Shivasubramanian Gopinath, Joseph Rosen, Vijayakumar Anand
https://arxiv.org/abs/2509.26179

Axial resolution post-processing engineering in Fresnel incoherent correlation holography
Fresnel incoherent correlation holography (FINCH) is a self-interference-based incoherent digital holography method. In FINCH, light from an object point is split into two beams, modulated differently using two lenses with different focal distances, and creates a self-interference hologram. At least three phase-shifted holograms are recorded and synthesized into a complex hologram, which reconstructs the object image without twin image and bias noises. Compared with conventional imaging, FINCH …

SafeEvalAgent: Toward Agentic and Self-Evolving Safety Evaluation of LLMs
Yixu Wang, Xin Wang, Yang Yao, Xinyuan Li, Yan Teng, Xingjun Ma, Yingchun Wang
https://arxiv.org/abs/2509.26100

SafeEvalAgent: Toward Agentic and Self-Evolving Safety Evaluation of LLMs
The rapid integration of Large Language Models (LLMs) into high-stakes domains necessitates reliable safety and compliance evaluation. However, existing static benchmarks are ill-equipped to address the dynamic nature of AI risks and evolving regulations, creating a critical safety gap. This paper introduces a new paradigm of agentic safety evaluation, reframing evaluation as a continuous and self-evolving process rather than a one-time audit. We then propose a novel multi-agent framework SafeE…

Self-driving cars: Are we there yet?
Merve Atasever, Zhuochen Liu, Qingpei Li, Akshay Hitendra Shah, Hans Walker, Jyotirmoy V. Deshmukh, Rahul Jain
https://arxiv.org/abs/2509.22754

Self-driving cars: Are we there yet?
Autonomous driving remains a highly active research domain that seeks to enable vehicles to perceive dynamic environments, predict the future trajectories of traffic agents such as vehicles, pedestrians, and cyclists and plan safe and efficient future motions. To advance the field, several competitive platforms and benchmarks have been established to provide standardized datasets and evaluation protocols. Among these, leaderboards by the CARLA organization and nuPlan and the Waymo Open Dataset …

Closure Relations of Synchrotron Self-Compton from Reverse shock and Fermi-LAT GRBs
Nissim Fraija, B. Betancourt Kamenetskaia, Antonio Galv\'an G\'amez, Maria G. Dainotti, Hermes Le\'on Vargas
https://arxiv.org/abs/2509.23345

Closure Relations of Synchrotron Self-Compton from Reverse shock and Fermi-LAT GRBs
Synchrotron radiation from the reverse- and forward-shock regions typically describes the evolution of temporal and spectral features given by the closure relations (CRs) during the late and long-/short-lasting emission in the afterglow phase of Gamma-ray bursts (GRBs). Although synchrotron photons are restricted to keV and a few MeV energies, the synchrotron self-Compton (SSC) mechanism can disperse them above hundreds of MeV energies. We present the CRs of the SSC process radiated from the re…

China's Global Digital Trade Expo in Hangzhou showcased robots, EVs, and a self-driving yacht, but questions remain over the US-China trade dispute and profits (Andrew Higgins/New York Times)
https://www.nytimes.com/2025/10/01/world/asia/china-tech-trade.html…

A Tech Expo Shows What China Can Make, but Not Who’ll Buy It All
The vast Global Digital Trade Expo in Hangzhou stood as a rebuke to U.S. efforts to hem in China’s technology. But the real competition is internal, and profits are hard to find.

Consciousness Self and Language
Robert Worden
https://arxiv.org/abs/2509.23342 https://arxiv.org/pdf/2509.23342

Consciousness Self and Language
Theories of consciousness depend on data, and it needs to be appropriate data, without overwhelming confounding factors. The reports of Minimal Phenomenal Experience (MPE) in [Metzinger 2024] relate to consciousness in a state purer than everyday consciousness, which may have fewer confounding factors. This essay suggests that the confounding factors, which are absent or diminished in MPE states, are related to language. The self which is absent in mindful states is a product of language. The l…

Non-local edge mode hybridization in the long-range interacting Kitaev chain
David Haink, Andreas A. Buchheit, Benedikt Fauseweh
https://arxiv.org/abs/2509.26447 https://…

Non-local edge mode hybridization in the long-range interacting Kitaev chain
In one-dimensional p-wave superconductors with short-range interactions, topologically protected Majorana modes emerge, whose mass decays exponentially with system size, as first shown by Kitaev. In this work, we extend this prototypical model by including power law long-range interactions within a self-consistent framework, leading to the self-consistent long-range Kitaev chain (seco-LRKC). In this model, the gap matrix acquires a rich structure where short-range superconducting correlations c…

A Scalable Framework for Safety Assurance of Self-Driving Vehicles based on Assurance 2.0
Shufeng Chen, Mariat James Elizebeth, Robab Aghazadeh Chakherlou, Xingyu Zhao, Eric Barbier, Siddartha Khastgir, Paul Jennings
https://arxiv.org/abs/2510.00092

A Scalable Framework for Safety Assurance of Self-Driving Vehicles based on Assurance 2.0
Assurance 2.0 is a modern framework developed to address the assurance challenges of increasingly complex, adaptive, and autonomous systems. Building on the traditional Claims-Argument-Evidence (CAE) model, it introduces reusable assurance theories and explicit counterarguments (defeaters) to enhance rigor, transparency, and adaptability. It supports continuous, incremental assurance, enabling innovation without compromising safety. However, limitations persist in confidence measurement, residu…

Self-hosting is becoming popular among a certain kind of user,
-- say the typical readership of ItsFoss.
There is a simple explanation for this shift:
people want their data, dollars, and destiny back.
Centralized platforms optimized for engagement and extraction are colliding with real-world needs
— privacy, compliance, predictability, and craft.
Linux, containers, and a flood of polished open-source apps have turned what used to be an enthusiast’s pro…

Self-Hosting is Rising and Linux Users are Leading This Revolution
Self‑hosting isn’t anti‑cloud; it’s pro‑agency. It’s choosing the right locus of control for the things you care about.

Geometric structure of stationary problem for spatial 1D self-diffusion equation with logistic growth
Yu ICHIDA
https://arxiv.org/abs/2509.24752 https://ar…

Geometric structure of stationary problem for spatial 1D self-diffusion equation with logistic growth
This paper considers the solution structure of non-trivial, non-constant stationary states of 1D spatial parabolic equations with nonlinear self-diffusion and logistic growth terms. A two-dimensional ordinary differential equation satisfying the stationary problem is derived and all its dynamics, including to infinity, is revealed by the Poincaré-Lyapunov compactification, one of the compactifications of phase space. The advantage of this method is that it can be used to classify all dynamical…

DeepProv: Behavioral Characterization and Repair of Neural Networks via Inference Provenance Graph Analysis
Firas Ben Hmida, Abderrahmen Amich, Ata Kaboudi, Birhanu Eshete
https://arxiv.org/abs/2509.26562

DeepProv: Behavioral Characterization and Repair of Neural Networks via Inference Provenance Graph Analysis
Deep neural networks (DNNs) are increasingly being deployed in high-stakes applications, from self-driving cars to biometric authentication. However, their unpredictable and unreliable behaviors in real-world settings require new approaches to characterize and ensure their reliability. This paper introduces DeepProv, a novel and customizable system designed to capture and characterize the runtime behavior of DNNs during inference by using their underlying graph structure. Inspired by system a…

Theory of Scaling Laws for In-Context Regression: Depth, Width, Context and Time
Blake Bordelon, Mary I. Letey, Cengiz Pehlevan
https://arxiv.org/abs/2510.01098 https://

Theory of Scaling Laws for In-Context Regression: Depth, Width, Context and Time
We study in-context learning (ICL) of linear regression in a deep linear self-attention model, characterizing how performance depends on various computational and statistical resources (width, depth, number of training steps, batch size and data per context). In a joint limit where data dimension, context length, and residual stream width scale proportionally, we analyze the limiting asymptotics for three ICL settings: (1) isotropic covariates and tasks (ISO), (2) fixed and structured covarianc…

Decentralized and Self-adaptive Core Maintenance on Temporal Graphs
Davide Rucci, Emanuele Carlini, Patrizio Dazzi, Hanna Kavalionak, Matteo Mordacchini
https://arxiv.org/abs/2510.00758

Decentralized and Self-adaptive Core Maintenance on Temporal Graphs
Key graph-based problems play a central role in understanding network topology and uncovering patterns of similarity in homogeneous and temporal data. Such patterns can be revealed by analyzing communities formed by nodes, which in turn can be effectively modeled through temporal $k$-cores. This paper introduces a novel decentralized and incremental algorithm for computing the core decomposition of temporal networks. Decentralized solutions leverage the ability of network nodes to communicate a…

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

On the Capacity of Self-Attention
Micah Adler
https://arxiv.org/abs/2509.22840 https://arxiv.org/pdf/2509.22840

On the Capacity of Self-Attention
While self-attention is known to learn relations among tokens, we lack a formal understanding of its capacity: how many distinct relations can a single layer reliably recover for a given budget? To formalize this, we introduce Relational Graph Recognition (RGR), where the key-query channel represents a graph on $m$ items with $m'$ directed edges, and, given a context of items, must recover the neighbors of each item. We measure resources by the total key dimension $D_K = h\,d_k$. Within this …

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization
Teng Zhang, Ziqian Fan, Mingxin Liu, Xin Zhang, Xudong Lu, Wentong Li, Yue Zhou, Yi Yu, Xiang Li, Junchi Yan, Xue Yang
https://arxiv.org/abs/2509.26281

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization
Driven by the growing need for Oriented Object Detection (OOD), learning from point annotations under a weakly-supervised framework has emerged as a promising alternative to costly and laborious manual labeling. In this paper, we discuss two deficiencies in existing point-supervised methods: inefficient utilization and poor quality of pseudo labels. Therefore, we present Point2RBox-v3. At the core are two principles: 1) Progressive Label Assignment (PLA). It dynamically estimates instance sizes…

Cowboys vs. Packers NFL player props: Self-Learning AI backing Jordan Love under 234.5 passing yards on SNF

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

Self-organized adaptive branching in frangible matter
P. L. B. Fischer, J. Tauber, T. Koch, L. Mahadevan
https://arxiv.org/abs/2509.26101 https://arxiv.org…

Self-organized adaptive branching in frangible matter
Soft and frangible materials that remodel under flow can give rise to branched patterns shaped by material properties, boundary conditions, and the time scales of forcing. We present a general theoretical framework for emergent branching in these frangible (or threshold) materials that switch abruptly from resisting flow to permitting flow once local stresses exceed a threshold, relevant for ex- amples as varied as dielectric breakdown of insulators and the erosion of soft materials. Simulation…

A minimal model of self-organized clusters with phase transitions in ecological communities
Shing Yan Li, Mehran Kardar, Zhijie Feng, Washington Taylor
https://arxiv.org/abs/2509.24985

A minimal model of self-organized clusters with phase transitions in ecological communities
In complex ecological communities, species may self-organize into clusters or clumps where highly similar species can coexist. The emergence of such species clusters can be captured by the interplay between neutral and niche theories. Based on the generalized Lotka-Volterra model of competition, we propose a minimal model for ecological communities in which the steady states contain self-organized clusters. In this model, species compete only with their neighbors in niche space through a common…

"Having Lunch Now": Understanding How Users Engage with a Proactive Agent for Daily Planning and Self-Reflection
Adnan Abbas, Caleb Wohn, Arnav Jagtap, Eugenia H Rho, Sang Won Lee
https://arxiv.org/abs/2509.24073

"Having Lunch Now": Understanding How Users Engage with a Proactive Agent for Daily Planning and Self-Reflection
Conversational agents have been studied as tools to scaffold planning and self-reflection for productivity and well-being. While prior work has demonstrated positive outcomes, we still lack a clear understanding of what drives these results and how users behave and communicate with agents that act as coaches rather than assistants. Such understanding is critical for designing interactions in which agents foster meaningful behavioral change. We conducted a 14-day longitudinal study with 12 parti…

Co-SOM: Co-training for photometric redshift estimation using Self-Organizing Maps
Alvaro Callejas-Tavera, Erik Molino-Minero-Re, Octavio Valenzuela
https://arxiv.org/abs/2509.25366

Co-SOM: Co-training for photometric redshift estimation using Self-Organizing Maps
The upcoming galaxy large-scale surveys, such as the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST), will generate photometry for billions of galaxies. The interpretation of large-scale weak lensing maps, as well as the estimation of galaxy clustering, requires reliable redshifts with high precision for multi-band photometry. However, obtaining spectroscopy for billions of galaxies is impractical and complex; therefore, having a sufficiently large number of galaxies with spe…

Stochastic Self-Organization in Multi-Agent Systems
Nurbek Tastan, Samuel Horvath, Karthik Nandakumar
https://arxiv.org/abs/2510.00685 https://arxiv.org/pd…

Stochastic Self-Organization in Multi-Agent Systems
Multi-agent systems (MAS) based on Large Language Models (LLMs) have the potential to solve tasks that are beyond the reach of any single LLM. However, this potential can only be realized when the collaboration mechanism between agents is optimized. Specifically, optimizing the communication structure between agents is critical for fruitful collaboration. Most existing approaches rely on fixed topologies, pretrained graph generators, optimization over edges, or employ external LLM judges, there…

Safety Instincts: LLMs Learn to Trust Their Internal Compass for Self-Defense
Guobin Shen, Dongcheng Zhao, Haibo Tong, Jindong Li, Feifei Zhao, Yi Zeng
https://arxiv.org/abs/2510.01088

Safety Instincts: LLMs Learn to Trust Their Internal Compass for Self-Defense
Ensuring Large Language Model (LLM) safety remains challenging due to the absence of universal standards and reliable content validators, making it difficult to obtain effective training signals. We discover that aligned models already possess robust internal safety beliefs: they consistently produce high-confidence refusals to harmful requests while exhibiting high entropy when generating potentially dangerous content. This entropy gap reveals an untapped signal--models intrinsically "know" wh…

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…

LatentEvolve: Self-Evolving Test-Time Scaling in Latent Space
Guibin Zhang, Fanci Meng, Guancheng Wan, Zherui Li, Kun Wang, Zhenfei Yin, Lei Bai, Shuicheng Yan
https://arxiv.org/abs/2509.24771

LatentEvolve: Self-Evolving Test-Time Scaling in Latent Space
Test-time Scaling (TTS) has been demonstrated to significantly enhance the reasoning capabilities of Large Language Models (LLMs) during the inference phase without altering model parameters. However, existing TTS methods are largely independent, implying that LLMs have not yet evolved to progressively learn how to scale more effectively. With the objective of evolving LLMs to learn ``how to scale test-time computation,'' we propose LatentEvolve, a self-evolving latent TTS framework inspired by…

Monday Night Football NFL player props: Self-learning AI backs Jake Browning Over 219.5 passing yards

https://www.cbssports.com/nfl/news/monday-

Probing phase transitions in non-Hermitian systems with quantum entanglement
Ling-Feng Zhang, Wing Chi Yu
https://arxiv.org/abs/2509.25703 https://arxiv.or…

Probing phase transitions in non-Hermitian systems with quantum entanglement
We study the quantum entanglement and quantum phase transition of the non-Hermitian anisotropic spin-1/2 XY model and XXZ model with the staggered imaginary field by analytical methods and numerical exact diagonalization, respectively. Various entanglement measures, including concurrence, negativity, mutual information, and quantum coherence, and both biorthogonal and self-normal quantities are investigated. Both the biorthogonal and self-normal entanglement quantities, except the biorthogonal …

Evaluating noises of boson sampling with statistical benchmark methods
Yang Ji, Yongjin Ye, Qiao Wang, Shi Wang, Jie Hou, Yongzheng Wu, Zijian Wang, Bo Jiang
https://arxiv.org/abs/2510.00056

Evaluating noises of boson sampling with statistical benchmark methods
The lack of self-correcting codes hiders the development of boson sampling to be large-scale and robust. Therefore, it is important to know the noise levels in order to cautiously demonstrate the quantum computational advantage or realize certain tasks. Based on those statistical benchmark methods such as the correlators and the clouds, which are initially proposed to discriminate boson sampling and other mockups, we quantificationally evaluate noises of photon partial distinguishability and ph…

One-DoF Robotic Design of Overconstrained Limbs with Energy-Efficient, Self-Collision-Free Motion
Yuping Gu, Bangchao Huang, Haoran Sun, Ronghan Xu, Jiayi Yin, Wei Zhang, Fang Wan, Jia Pan, Chaoyang Song
https://arxiv.org/abs/2509.22002

One-DoF Robotic Design of Overconstrained Limbs with Energy-Efficient, Self-Collision-Free Motion
While it is expected to build robotic limbs with multiple degrees of freedom (DoF) inspired by nature, a single DoF design remains fundamental, providing benefits that include, but are not limited to, simplicity, robustness, cost-effectiveness, and efficiency. Mechanisms, especially those with multiple links and revolute joints connected in closed loops, play an enabling factor in introducing motion diversity for 1-DoF systems, which are usually constrained by self-collision during a full-cycle…

Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports
Changde Du, Yizhuo Lu, Zhongyu Huang, Yi Sun, Zisen Zhou, Shaozheng Qin, Huiguang He
https://arxiv.org/abs/2509.24298

Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports
The ability to represent emotion plays a significant role in human cognition and social interaction, yet the high-dimensional geometry of this affective space and its neural underpinnings remain debated. A key challenge, the `behavior-neural gap,' is the limited ability of human self-reports to predict brain activity. Here we test the hypothesis that this gap arises from the constraints of traditional rating scales and that large-scale similarity judgments can more faithfully capture the brain'…

Learning the Neighborhood: Contrast-Free Multimodal Self-Supervised Molecular Graph Pretraining
Boshra Ariguib, Mathias Niepert, Andrei Manolache
https://arxiv.org/abs/2509.22468

Learning the Neighborhood: Contrast-Free Multimodal Self-Supervised Molecular Graph Pretraining
High-quality molecular representations are essential for property prediction and molecular design, yet large labeled datasets remain scarce. While self-supervised pretraining on molecular graphs has shown promise, many existing approaches either depend on hand-crafted augmentations or complex generative objectives, and often rely solely on 2D topology, leaving valuable 3D structural information underutilized. To address this gap, we introduce C-FREE (Contrast-Free Representation learning on Ego…

Sonnet 040 - XL
Take all my loves, my love, yea take them all;
What hast thou then more than thou hadst before?
No love, my love, that thou mayst true love call;
All mine was thine, before thou hadst this more.
Then, if for my love, thou my love receivest,
I cannot blame thee, for my love thou usest;
But yet be blam'd, if thou thy self deceivest
By wilful taste of what thyself refusest.
I do forgive thy robbery, gentle thief,
Al…

Pony.ai becomes the first company to receive a citywide robotaxi permit for Shenzhen, allowing self-driving taxis to operate beyond pilot zones (Evelyn Cheng/CNBC)
https://www.cnbc.com/2025/10/31/chinas-ponyai-gets-the-first-permit-fo…

Pony.ai becomes first to win citywide robotaxi permit in China's Silicon Valley
Pony.ai said it is the first company to get a permit to operate taxis across all of the Chinese city of Shenzhen on the border of Hong Kong.

Geometric Frustration Directs the Self-Assembly of Nanoparticles with Crystallized Ligand Bundles
Federico Tomazic, Aswathy Muttathukattil, Afshin Nabiyan, Felix Schacher, Michael Engel
https://arxiv.org/abs/2509.23337

Geometric Frustration Directs the Self-Assembly of Nanoparticles with Crystallized Ligand Bundles
Polymer-grafted nanoparticles are versatile building blocks that self-assemble into a rich diversity of mesostructures. Coarse-grained molecular simulations have commonly accompanied experiments by resolving structure formation pathways and predicting phase behavior. Past simulations represented nanoparticles as spheres and the ligands as flexible chains of beads, isotropically tethered to the nanoparticles. Here, we investigate a different minimal coarse-grained model. The model consists of an…

Self-learning AI reveals NFL against the spread, over/under, money-line picks for every Week 5, 2025 game

https://www.cbssports.com/nfl/news/ai-nfl-picks-against-spread…

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

CCNeXt: An Effective Self-Supervised Stereo Depth Estimation Approach
Alexandre Lopes, Roberto Souza, Helio Pedrini
https://arxiv.org/abs/2509.22627 https://

CCNeXt: An Effective Self-Supervised Stereo Depth Estimation Approach
Depth Estimation plays a crucial role in recent applications in robotics, autonomous vehicles, and augmented reality. These scenarios commonly operate under constraints imposed by computational power. Stereo image pairs offer an effective solution for depth estimation since it only needs to estimate the disparity of pixels in image pairs to determine the depth in a known rectified system. Due to the difficulty in acquiring reliable ground-truth depth data across diverse scenarios, self-supervis…

TASP: Topology-aware Sequence Parallelism
Yida Wang (Capital Normal University, Infinigence-AI), Ke Hong (Tsinghua University, Infinigence-AI), Xiuhong Li (Infinigence-AI), Yuanchao Xu (Capital Normal University), Wenxun Wang (Tsinghua University), Guohao Dai (Infinigence-AI, Shanghai Jiao Tong University), Yu Wang (Tsinghua University)
https://

TASP: Topology-aware Sequence Parallelism
Long-context large language models (LLMs) face constraints due to the quadratic complexity of the self-attention mechanism. The mainstream sequence parallelism (SP) method, Ring Attention, attempts to solve this by distributing the query into multiple query chunks across accelerators and enable each Q tensor to access all KV tensors from other accelerators via the Ring AllGather communication primitive. However, it exhibits low communication efficiency, restricting its practical applicability. …

Inclusive Easy-to-Read Generation for Individuals with Cognitive Impairments
Fran\c{c}ois Ledoyen, Ga\"el Dias, Alexis Lechervy, Jeremie Pantin, Fabrice Maurel, Youssef Chahir, Elisa Gouzonnat, M\'elanie Berthelot, Stanislas Moravac, Armony Altinier, Amy Khairalla
https://arxiv.org/abs/2510.00691

Inclusive Easy-to-Read Generation for Individuals with Cognitive Impairments
Ensuring accessibility for individuals with cognitive impairments is essential for autonomy, self-determination, and full citizenship. However, manual Easy-to-Read (ETR) text adaptations are slow, costly, and difficult to scale, limiting access to crucial information in healthcare, education, and civic life. AI-driven ETR generation offers a scalable solution but faces key challenges, including dataset scarcity, domain adaptation, and balancing lightweight learning of Large Language Models (LLM…

JEPA-T: Joint-Embedding Predictive Architecture with Text Fusion for Image Generation
Siheng Wan, Zhengtao Yao, Zhengdao Li, Junhao Dong, Yanshu Li, Yikai Li, Linshan Li, Haoyan Xu, Yijiang Li, Zhikang Dong, Huacan Wang, Jifeng Shen
https://arxiv.org/abs/2510.00974

JEPA-T: Joint-Embedding Predictive Architecture with Text Fusion for Image Generation
Modern Text-to-Image (T2I) generation increasingly relies on token-centric architectures that are trained with self-supervision, yet effectively fusing text with visual tokens remains a challenge. We propose \textbf{JEPA-T}, a unified multimodal framework that encodes images and captions into discrete visual and textual tokens, processed by a joint-embedding predictive Transformer. To enhance fusion, we incorporate cross-attention after the feature predictor for conditional denoising while main…

Rethinking Thinking Tokens: LLMs as Improvement Operators
Lovish Madaan, Aniket Didolkar, Suchin Gururangan, John Quan, Ruan Silva, Ruslan Salakhutdinov, Manzil Zaheer, Sanjeev Arora, Anirudh Goyal
https://arxiv.org/abs/2510.01123

Rethinking Thinking Tokens: LLMs as Improvement Operators
Reasoning training incentivizes LLMs to produce long chains of thought (long CoT), which among other things, allows them to explore solution strategies with self-checking. This results in higher accuracy, but inflates context length, token/compute cost, and answer latency. We ask: Can current models leverage their metacognition to provide other combinations on this Pareto frontier, e.g., better accuracy with lower context length and/or latency? Abstractly, we view the model as an improvement op…

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[2/7]:
- Internalizing Self-Consistency in Language Models: Multi-Agent Consensus Alignment
Samanta, Magesh, Yu, Wu, Jain, Jiang, Vidolov, Sajda, Efroni, Hassani

Optimizing Speech Language Models for Acoustic Consistency
Morteza Rohanian, Michael Krauthammer
https://arxiv.org/abs/2509.26276 https://arxiv.org/pdf/250…

Optimizing Speech Language Models for Acoustic Consistency
We study speech language models that incorporate semantic initialization and planning losses to achieve robust and consistent generation. Our approach initializes speech tokens with self-supervised features, applies a light alignment loss, and trains with thinning and auxiliary objectives that target robustness and content planning. We train three models: a 0.7B speech-only model, a 1.0B speech-only model, and a 1.0B interleaved model with both text and speech. Acoustic studies show that the sp…

A Neuro-Fuzzy System for Interpretable Long-Term Stock Market Forecasting
Miha O\v{z}bot, Igor \v{S}krjanc, Vitomir \v{S}truc
https://arxiv.org/abs/2510.00960 https://

A Neuro-Fuzzy System for Interpretable Long-Term Stock Market Forecasting
In the complex landscape of multivariate time series forecasting, achieving both accuracy and interpretability remains a significant challenge. This paper introduces the Fuzzy Transformer (Fuzzformer), a novel recurrent neural network architecture combined with multi-head self-attention and fuzzy inference systems to analyze multivariate stock market data and conduct long-term time series forecasting. The method leverages LSTM networks and temporal attention to condense multivariate data into i…

Scaling Spoken Language Models with Syllabic Speech Tokenization
Nicholas Lee, Cheol Jun Cho, Alan W Black, Gopala K. Anumanchipalli
https://arxiv.org/abs/2509.26634 https://

Scaling Spoken Language Models with Syllabic Speech Tokenization
Spoken language models (SLMs) typically discretize speech into high-frame-rate tokens extracted from SSL speech models. As the most successful LMs are based on the Transformer architecture, processing these long token streams with self-attention is expensive, as attention scales quadratically with sequence length. A recent SSL work introduces acoustic tokenization of speech at the syllable level, which is more interpretable and potentially more scalable with significant compression in token len…