arXiv preprints, newest first, from January 1, 2026 through July 17, 2026.
Search runs only over papers downloaded into this site's SQLite database.
Multimodal large language models have achieved strong performance across diverse vision-language tasks, yet their capabilities in UAV scenarios remain insufficiently explored. Recent UAV-oriented benchmarks have begun to evaluate MLLMs in aerial scenarios, but they typically focus on scene understanding, event recognition, or...
Humans can infer how objects are likely to move from passive observation: a cup may be lifted, a drawer may slide, and a lid may rotate shut. Such predictions expose the physical consequences of interaction needed to act in the real world. We study how to learn this anticipation from ordinary monocular videos of human-object...
Interactions between dark matter and dark energy can lead to an apparent phantom-crossing behavior that mimics the expansion history preferred by the latest cosmological observations from DESI baryon acoustic oscillations (BAO), Cosmic Microwave Background (CMB), and Type Ia supernovae (SNe Ia) data. In a previous paper [Khoury, Lin,...
Video Diffusion Transformers process long spatio-temporal sequences, making self-attention the main bottleneck in high-resolution video generation. Training-free sparse attention reduces this cost, but adaptive Top-$p$ routing creates uneven per-head workloads under multi-GPU sequence parallelism. The resulting workload heterogeneity...
Grounded long-video question answering (Grounded LVQA) requires answering a question about a long video while localizing the short evidence interval that supports the answer. Recent agentic methods frame this task as multi-turn exploration with a single crop_video(start, end) action, which supports coarse-to-fine narrowing but...
We study the problem of optimal control for dynamical systems near a pitchfork bifurcation, motivated by the role of external cues in guiding symmetry-breaking transitions in cell-fate selection and other natural processes. Using an asymptotic expansion of the optimality conditions obtained from the Pontryagin maximum principle, the...
Dexterous hands and humanoid robots are typically developed as distinct embodiments: the former enable contact-rich manipulation at the object scale, whereas the latter provide mobility and whole-body interaction in human-centered environments. We introduce \textbf{Handroid}, a desktop-scale dual-embodiment robot that integrates both...
Maxwell demons can convert knowledge into thermodynamic advantage by using measurement-acquired information. However, standard formulations of this problem assume that the demon has access to the system's entire phase space, an assumption that fails in many practical applications. Here, I address this problem by deriving fluctuation...
We study $(2+1)$-dimensional oscillatory integral operators of the form \[ T_λf(x,y)=\int_{\mathbb{R}}e^{iλP(x,y)t^k}ψ(x,y,t)f(t)dt,\qquad k\geq 1, \] where the phase $P$ is a real-analytic function with a critical point at the origin. We establish the sharp $L^2\to L^2$ decay rate of $\frac12\min\{1/h_{P}, 1/k\}$, where $h_{P}$...
Mixture-of-Experts (MoE) is a popular class of large language models (LLMs), offering high efficiency and accuracy. However, in KV-cache-intensive serving scenarios, MoEs often exhibit a tension between the GPU memory requirements of the model weights and the growing KV cache. We propose PagedWeight, a novel management method for MoE...
Geometric scaling was conjectured and observed at the ISR more than 50 years ago. We argue that it still holds at the LHC. We show that the dip-bump structures of the differential elastic cross-sections exhibit a remarkable regularity, namely that the ratio of the bump to dip positions is constant from 23 GeV to 13 TeV. Using crossing...
To address the escalating energy and latency demands of machine-learning workloads, we introduce a blueprint for an energy-efficient and fast thermodynamic computing stack that leverages stochastic analog processes in physical hardware. In this work, we focus on energy-based thermodynamic computing where the stochastic process is well...
This study introduces a vision-language pipeline that detects risky driving behaviors and generates emotionally expressive responses to support driver awareness and comfort. Although vision-language models have advanced perception and reasoning in autonomous driving, existing systems rarely consider the emotional dimension or...
Ferroelectric materials are promising platforms for controllable photonic devices because of the strong coupling between their spontaneous polarization and optical properties. Yet, these materials remain challenging to design because of the close connection between the ferroelectric domain structure and optical response, which no...
Inspired by Wigner's classification of elementary particles as irreducible unitary representations of a spacetime symmetry group, we ask what group can give rise in this way to the quantum states of a particle dressed with clouds of infrared gauge bosons. We show that the answer is given by standard asymptotic symmetries (such as BMS...
In many applications of matching, the point clouds to be matched are not merely unstructured sets of points but rather samples from distributions with an intrinsic cluster structure. In such cases, as individual points are often interchangeable within a coherent region, finding a robust region-to-region alignment is more desirable...
Reinforcement learning (RL) has recently emerged as a promising feedback control strategy for nonlinear and complex dynamical systems. However, RL algorithms are sample inefficient and require a large number of interaction with the environment to synthesize optimal control strategies. Consequently, applications of RL are typically...
Synthetic optical holography (SOH) introduced holographic reconstruction into scanning optical microscopy, enabling quantitative phase imaging with sequential acquisition and point detection. Here, we extend this concept to the quantum regime by implementing SOH within quantum imaging with undetected light (QIUL). By integrating a...
Connected and Autonomous Vehicles (CAVs) rely on interconnected software and hardware components, including sensors, Electronic Control Units, in-vehicle infotainment systems, and telematics units, where vulnerabilities can compromise assets, users, and vehicle operations. These vulnerabilities are commonly documented as plain text in...
Neutron-rich Si isotopes represent a unique case of shell evolution, exhibiting a robust shell closure at $N=20$ and pronounced quadrupole collectivity at $N = 28$. We report lifetime measurements of excited states in $^{40}$Si and the first simultaneous lifetime and heavy-ion inelastic-scattering measurements in $^{41}$Si. In...
Open-vocabulary 3D maps let robots answer language queries about what and where, but they assume a static world and cannot answer queries about how scene elements behave. We introduce Vision-Language-Motion Maps (VLMM), an open-vocabulary, natural-language-queryable 3D map in which each element carries a fused motion attribute: a...
The modern manufacturing industry is investing in new technologies such as the Internet of Things (IoT), big data analytics, cloud computing and cybersecurity to cope with system complexity, increase information visibility, improve production performance, and gain competitive advantages in the global market. These advances are rapidly...
We disprove the conjecture that every globally asymptotically stable homogeneous polynomial vector field admits a homogeneous polynomial Lyapunov function. The counterexample is a planar homogeneous cubic polynomial vector field with integer coefficients. It admits no positive definite homogeneous polynomial with nonpositive Lie...
The optical morphologies of low-mass galaxies can be used to directly trace their assembly and constrain models of galaxy evolution. We select a sample of 6211 low-mass ($7\lesssim {\rm log}(M_{\ast}/M_{\odot})\lesssim 10$) star-forming galaxies from the SAGAbg catalog that has high-completeness at low-redshifts ($z<0.1$). We obtain...
Muon is competitive with AdamW in large-scale pre-training, but its value for reinforcement-learning (RL) post-training remains unclear. We study vanilla Muon in sparse-reward agentic RL through matched single-seed comparisons with AdamW on ALFWorld using Qwen2.5-0.5B-Instruct. Under Group-in-Group Policy Optimization (GiGPO),...