arXiv preprints from January 1, 2026 through September 23, 2026 — 03:34:06 EST
Accounting for information flow on the path space of trajectories of a nonnegative martingale yields exact variational identities for it, even at arbitrary random times. This recovers the widely used classical concentration inequalities, from Ville to PAC-Bayes, and measures what each one discards. The tail a bound controls is itself...
Large Language Models (LLMs) increasingly require selective removal of harmful or sensitive knowledge, called unlearning, yet existing methods and benchmarks fail to evaluate this capability completely. Current approaches rely on disjoint forget and retain sets composed of independent facts, and measure success using simple and direct...
We present 4DAnyone, a framework for reconstructing 4D humans from an uncalibrated monocular video by generating reconstruction-grade multiview-consistent videos and lifting them into 4D Gaussian Splatting (4DGS). Existing camera-controlled video diffusion models synthesize plausible novel-view videos but fail to maintain consistency...
Identity-preserving image generation becomes increasingly unreliable when a scene must contain many specified people. Beyond retaining each identity, the model must bind every reference to a distinct person and location, while training-time identity losses must establish correspondence among several noisy predicted faces. We introduce...
We present Swift-Image, a compact unified model for text-to-image generation, single-image editing, and multi-image editing. Our goal is to explore how far a relatively small visual generator can be pushed through systematic training engineering under a constrained computational budget. Swift-Image adopts an efficient 6B single-stream...
For an untwisted Wess--Zumino--Witten modular tensor category $\cC(\mathfrak g,k)$, let $r(\mathfrak g,k)$ be the number of simple objects and $\cD(\mathfrak g,k)$ its total quantum dimension. The vacuum-flux state on a torus divided into two cylinders has positive universal entropy contribution $Γ_{T^2}=2\log\cD$. We maximize this...
We propose the high-fidelity preparation of single bosonic molecules in optical tweezers starting from small tweezer-trapped molecular ensembles. Our scheme combines a static electric field and a microwave field to generate strong, tunable, anisotropic interactions that shield the molecules against two-body collisional loss. We show...
Personalized interpretation of medical reports has emerged as an increasingly important need among patients. Addressing this need requires both evidence-grounded medical factuality and context-dependent patient communication, yet existing medical vision-language tasks do not adequately capture these dual requirements. To bridge this...
We show that for Abelian Chern-Simons topological orders, torus Berry matrices determine the all-genus extended TQFT. We identify the topological part of the projective Berry holonomy under metric deformations with the mapping-class-group representation of the Abelian Chern-Simons TQFT and prove that the normalized torus data...
In this manuscript, we analyze the contribution of fragmentation mechanisms to the inclusive hadroproduction of heavy quarkonium pairs, taking into account both single- and double-hadron (dihadron) fragmentation channels. To estimate the contribution of the latter mechanism, we construct a microscopic perturbative model of the...
Modern experiments often record nonequilibrium dynamics as sequences of discrete events whose rates depend on the realized past. We develop a fluctuation-response theory for such non-Markovian jump processes directly on the observed event record. Memory can destroy a closed master equation for the state probabilities. Each transition...
Supersolids are exotic quantum states in which long-range phase coherence coexists, and may interplay, with emergent spatial orders. A particularly rich phase diagram featuring several competing spatial orders is predicted for dipolar supersolids with two-dimensional crystals, yet the experimental observation of this structural...
Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions. Consequently, users lack a reliable signal for deciding when a prediction can be trusted. Post-hoc confidence estimation addresses this by training a lightweight auxiliary head over a frozen classifier. Existing targets, however,...
The unusual transport properties of strange metals point to the breakdown of the quasiparticle picture, the understanding of which remains one of the most vexing problems in physics. Here, we report investigations of the electrical Hall effect of the strange metal superconductor BaFe$_2$(As$_{1-x}$P$_x$)$_2$. We show that a...
The choice of hydrodynamic frame directly influences the numerical values of transport coefficients in relativistic dissipative hydrodynamics. We derive the exact transformation between the Eckart and Landau--Lifshitz frames and show that their thermal conductivities are related by an enthalpy-dependent factor. Using a baryon-rich...
We analyze the performance of a spin-torque microwave detector (STMD) driven by positive rectangular current pulses $I(t)$ of various amplitudes $I_0$, durations $τ$, and repetition periods $T$ and reveal two distinct regimes of STMD operation. In the first (linear) regime, the time-averaged voltage across the detector, $U_{\rm dc}$,...
Despite their growing importance for contact-free radio frequency (RF) based healthcare monitoring, different radio technologies such as frequency-modulated continuous wave (FMCW) radar, impulse radio ultra-wideband (IR-UWB), and Wi-Fi sensing are rarely compared under identical deployment conditions, as existing studies typically...
We study large-sample properties of higher-order Markov chains on a finite alphabet $Σ$ when the order $m_n$ is allowed to grow with the sequence length $n$. By embedding the process into a first-order chain on $Σ^{m_n}$ and exploiting return-time decompositions, we establish a central limit theorem for additive functionals...
Travel behavior research increasingly combines digital data collection with predictive modeling, yet these stages are often developed and evaluated separately. This study proposes a three-agent workflow integrating conversational data collection, structured data processing, and behavioral prediction. A chatbot-administered,...
Naturalistic computer-use traces, passively recorded screenshots and mouse or keyboard actions, are a valuable resource for deriving symbolic, auditable, and reusable models of how everyday work is done. Such models matter as computer-use agents enter real work, where agents need to learn how tasks are actually performed, and...
Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the training algorithm: a better objective or update rule improves the compute\mbox{-}capability exchange rate for every subsequent run, including the one that...
The BrowseComp-Plus benchmark disentangled the evaluation of agentic search by replacing opaque web search with a fixed corpus, so that an agent's role can be separated from the retriever's. That corpus, however, holds only about 100K documents and was assembled from the supporting documents of the benchmark's own queries plus mined...
Heterogeneous AI systems composed of multiple models, architectures, harnesses, or inference-time settings can improve quality and efficiency by routing queries to the specialist who can answer most effectively at the lowest cost. Routing requires estimating each specialist's expected return, but this value estimation has a cost....
A single training example's contribution to a finished model is normally estimated rather than measured, because measuring it takes two expensive full pre-training runs that differ in one row of one batch. We ran that counterfactual 24 times at a small scale. We trained 32 GPT-2 models at 124M parameters from scratch on OpenWebText,...
We prove matching upper and lower bounds on the space complexity of simulating a large shared register using smaller shared registers. We focus on the case where both the simulated and base registers are single-writer, which means they can be accessed concurrently by multiple readers but only by a single writer. To strengthen our...