arXiv preprints from January 1, 2026 through September 26, 2026 — 01:17:25 EST
We report a case study of four end-to-end attempts to autonomously generate ML research papers using a pipeline of six LLM agents mapped to stages of the scientific workflow. Of these four, three attempts failed during implementation or evaluation. One completed the pipeline and was accepted to Agents4Science 2025, an experimental...
Large language models (LLMs) excel in program synthesis, yet their capacity for neural architecture design -- balancing syntactic reliability, performance, and structural novelty -- remains underexplored. We present a closed-loop architecture synthesis pipeline within the NNGPT framework, in which a code-oriented LLM evolves over 22...
Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attributed to their capability to generate explicit chain-of-thought (CoT) explanations. However, recent work shows that LRMs often arrive at the correct answer before completing these textual reasoning steps, indicating the presence of...
In our previous work~\cite{SanchisAgudoVinuesa2025PRL}, we argued that viscous dissipation in turbulence can be understood as the macroscopic imprint of microscopic path uncertainty, and showed that a kernel variance field $s(y)$ constrained by a balance condition yields both the Kolmogorov scales and the logarithmic law of the wall...
We present new Rosseland and Planck mean opacity tables relevant to the shallow interiors and atmospheres of giant planets. The tables span metallicities from 0.31 to 50 times solar, temperatures from 100 - 6000 K, and pressures from 1e-6 - 1e5 bar, thereby covering a wider parameter space than previous data sets. Our calculations...
Vision-based robotic policies often struggle with even minor viewpoint changes, underscoring the need for view-invariant visual representations. This challenge becomes more pronounced in real-world settings, where viewpoint variability is unavoidable and can significantly disrupt policy performance. Existing methods typically learn...
Retrieval-Augmented Generation (RAG) has become a key paradigm for reducing factual hallucinations in Large Language Models (LLMs), yet little is known about how the order of retrieved documents affects model behavior. We empirically show that under a Top-5 retrieval setting with the gold document included, LLM answers vary...
Lawler and Trujillo Ferreras constructed a well-known coupling between the Brownian loop soups on $\mathbb{R}^2$ and the (discrete-time) random walk loop soups on $\mathbb{Z}^2$ (one rescales the random walk loops by $1/N$, their time parametrizations by $1/(2N^2)$, and lets $N\to \infty$), which led to numerous applications. It...
Comic understanding presents a significant challenge for Multimodal Large Language Models (MLLMs), as the intended meaning of a comic often emerges from the joint interpretation of visual, textual, and social cues. This naturally motivates Chain-of-Thought (CoT) prompting, since explicit intermediate reasoning appears promising for...
We study the neutrinoless double beta decays of spin-1/2 hyperons, $B_1^- \to B_2^+ \ell^- \ell^-$ with $B_1^-\in \{Σ^-,Ξ^-\}$ and $B_2^+\in\{p,Σ^+\}$, which violate lepton number by two units. The decay amplitudes are computed within covariant baryon chiral perturbation theory at the one-loop level. Dalitz plots of the squared...
Large language models (LLMs), despite strong performance on complex mathematical problems, exhibit systematic limitations in counting tasks. This issue arises from the architectural limits of transformers, where counting is performed across layers, leading to degraded precision for larger counting problems due to depth constraints. To...
In this study, we present ULS+, an enhanced version of the Universal Lesion Segmentation (ULS) model. The original ULS model segments lesions across the whole body in CT scans given volumes of interest (VOIs) centered around a click-point. Since its release, several new public datasets have become available that can further improve...
Text-to-Image editing using diffusion models faces challenges in balancing content preservation with edit application and handling real-image editing. To address these, we propose LAMS-Edit, leveraging intermediate states from the inversion process--an essential step in real-image editing--during edited image generation. Specifically,...
Recent approaches in personalized reward modeling have primarily focused on leveraging user interaction history to align model judgments with individual preferences. However, existing approaches largely treat user context as a static or implicit conditioning signal, failing to capture the dynamic and multi-faceted nature of human...
An in-principle exact working equation to compute electronic affinity and ionization Fukui functions is derived within the $N$-centered (Nc) ensemble extension of density functional theory (DFT). It circumvents the kernel derivative discontinuity problem of DFT for fractional electron numbers, whose contribution is recovered through...
The aim of this work is to enhance blockchain security by deepening the understanding of selfish mining attacks in various consensus protocols, especially the ones that have the potential to mitigate selfish mining. Previous research was mainly focused on a particular protocol with a single selfish miner, while only limited studies...
Autonomous systems (AS) powered by AI components are increasingly integrated into the fabric of our daily lives and society, raising concerns about their ethical and social impact. To be considered trustworthy, AS must adhere to ethical principles and values. This has led to significant research on the identification and incorporation...
Recent advances in audio large language models (ALLMs) have made high-quality synthetic audio widely accessible, increasing the risk of malicious audio deepfakes across speech, environmental sounds, singing voice, and music. Real-world audio deepfake detection (ADD) therefore requires all-type detectors that generalize across...
In this paper, we study two inflationary models, namely, monomial inflation and the simplest $α$-attractor inflation, within extra-dimensional frameworks. We consider three extra-dimensional setups: Dark Dimension, which embeds one flat extra-dimension to explain the observed smallness of the 4D cosmological constant $Λ_4$; and the...
Prompt tuning has achieved remarkable progress in vision-language models (VLMs) and is recently being adopted for audio-language models (ALMs). However, its generalization ability in ALMs remains largely underexplored. We observe that conventional prompt tuning for ALMs also suffers from the Base-New Tradeoff, and we identify that...
The high rate of development of Internet of Things (IoT) devices has brought to attention new challenges in the area of data security, especially within the resource-limited realm of RFID tags, sensors, and embedded systems. Traditional cryptographic implementations can be of inappropriate computational complexity and energy usage and...
The equation-of-state (EOS) parameter $φ\equiv P/\varepsilon$, defined as the ratio of pressure to energy density, encapsulates the fundamental response of matter under extreme compression. Its value at the center of the most massive neutron star (NS), $\x \equiv φ_{\rm c} = P_{\rm c}/\varepsilon_{\rm c}$, sets a universal upper bound...
For every flat surface, almost every flat surface in its $\mathsf{SL}(2,\mathbb{R})$ orbit has the following property: the sequence of its saddle connection lengths in non-decreasing order is uniformly distributed in the unit interval.
Most Integrated Sensing and Communications (ISAC) systems require dividing airtime across their two modes. However, the specific impact of this decision on sensing performance remains unclear and underexplored. In this paper, we therefore investigate the impact on a gesture recognition system using a Millimeter-Wave (mmWave) ISAC...
Recent work in Mechanistic Interpretability (MI) has enabled the identification and intervention of internal features in Large Language Models (LLMs). However, a persistent challenge lies in linking such internal features to the reliable control of complex, behavior-level semantic attributes in language generation. In this paper, we...