arXiv subjects

:

Publications and source records attributed to :.

At least 163 records · Page 9Linked to original sources

arXiv2025

Model-independent measurement of $D^0$-$\overline{D}{}^0$ mixing parameters in $D^0\rightarrow K^0_{S}π^+π^-$ decays at Belle and Belle II

We perform a model-independent measurement of the $D^0$-$\overline{D}{}^0$ mixing parameters using samples of $e^+e^-$-collision data collected by the Belle and Belle II experiments that have integrated luminosities of $951\ \text{fb}^{-1}$ and $408\ \text{fb}^{-1}$, respectively. Approximately $2.05\times10^6$ neutral $D$ mesons are reconstructed in the $D^0\rightarrow K^0_{S}π^+π^-$ channel, with the neutral $D$ flavor tagged by the charge of the pion in the $D^{*+}\rightarrow D^0π^+$ decay. Assuming charge-parity symmetry, the mixing parameters are measured to be $ x = (4.0\pm1.7\pm0.4)\times 10^{-3} $ and $ y = (2.9\pm1.4\pm0.3)\times 10^{-3}$, where the first uncertainties are statistical and the second systematic. The results are consistent with previous determinations.

hep-ex↗

arXiv2025

Deep Research Bench: Evaluating AI Web Research Agents

Amongst the most common use cases of modern AI is LLM chat with web search enabled. However, no direct evaluations of the quality of web research agents exist that control for the continually-changing web. We introduce Deep Research Bench, consisting of 89 multi-step web research task instances of varying difficulty across 8 diverse task categories, with the answers carefully worked out by skilled humans. We provide a "RetroSearch" environment with a large frozen set of scraped web pages, and demonstrate that offline "RetroSearch" agents perform comparably to "live web" agents, enabling reliable evaluations of models over time. We provide robust agent tooling and scaffolding to benchmark major LLMs as they are released, including "thinking" models like o3 and Gemini 2.5 Pro. We include automated evaluations of the lengthy agent traces to report progress over time in hallucinations, tool use, and forgetting. Finally, we evaluate the major web research products branded as "Deep Research", "Deep Search", "Search", or "Research." Results are available on a public leaderboard at https://drb.futuresearch.ai/.

cs.AI↗

arXiv2025

Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning

We introduce Seed1.5-Thinking, capable of reasoning through thinking before responding, resulting in improved performance on a wide range of benchmarks. Seed1.5-Thinking achieves 86.7 on AIME 2024, 55.0 on Codeforces and 77.3 on GPQA, demonstrating excellent reasoning abilities in STEM and coding. Beyond reasoning tasks, the method demonstrates notable generalization across diverse domains. For instance, it surpasses DeepSeek R1 by 8% in win rate on non-reasoning tasks, indicating its broader applicability. Compared to other state-of-the-art reasoning models, Seed1.5-Thinking is a Mixture-of-Experts (MoE) model with a relatively small size, featuring 20B activated and 200B total parameters. As part of our effort to assess generalized reasoning, we develop two internal benchmarks, BeyondAIME and Codeforces, both of which will be publicly released to support future research. Model trial link: https://www.volcengine.com/experience/ark.

cs.CL↗

arXiv2025

Helicity-dependent parton distribution functions at next-to-next-to-leading order accuracy from inclusive and semi-inclusive deep-inelastic scattering data

We present MAPPDFpol1.0, a new determination of the helicity-dependent parton distribution functions (PDFs) of the proton from a set of longitudinally polarised inclusive and semi-inclusive deep-inelastic scattering data. The determination includes, for the first time, next-to-next-to-leading order QCD corrections to both processes, and is carried out in a framework that combines a neural-network parametrisation of PDFs with a Monte Carlo representation of their uncertainties. We discuss the quality of the determination, in particular its dependence on higher-order corrections, on the choice of data set, and on theoretical constraints.

hep-ph↗

arXiv2025

Search for lepton-flavor-violating $τ^- \to \ell^- K_s^0$ decays at Belle and Belle II

We present the results of a search for charged-lepton-flavor violating decays $τ^{-} \rightarrow \ell^{-}K_{S}^{0}$, where $\ell^{-}$ is either an electron or a muon. We combine $e^+e^-$ data samples recorded by the Belle II experiment at the SuperKEKB collider (428 fb$^{-1}$) with samples recorded by the Belle experiment at the KEKB collider (980 fb$^{-1}$) to obtain a sample of 1.3 billion $e^+e^-\toτ^+τ^-$ events. We observe 0 and 1 events and set $90\%$ confidence level upper limits of $0.8 \times 10^{-8}$ and $1.2 \times 10^{-8}$ on the branching fractions of the decay modes $τ^{-} \rightarrow e^{-}K_{S}^{0}$ and $τ^{-} \rightarrow μ^{-}K_{S}^{0}$, respectively. These are the most stringent upper limits to date.

hep-ex↗

arXiv2025

Command A: An Enterprise-Ready Large Language Model

In this report we describe the development of Command A, a powerful large language model purpose-built to excel at real-world enterprise use cases. Command A is an agent-optimised and multilingual-capable model, with support for 23 languages of global business, and a novel hybrid architecture balancing efficiency with top of the range performance. It offers best-in-class Retrieval Augmented Generation (RAG) capabilities with grounding and tool use to automate sophisticated business processes. These abilities are achieved through a decentralised training approach, including self-refinement algorithms and model merging techniques. We also include results for Command R7B which shares capability and architectural similarities to Command A. Weights for both models have been released for research purposes. This technical report details our original training pipeline and presents an extensive evaluation of our models across a suite of enterprise-relevant tasks and public benchmarks, demonstrating excellent performance and efficiency.

cs.CL↗

arXiv2025

Rethinking Reflection in Pre-Training

A language model's ability to reflect on its own reasoning provides a key advantage for solving complex problems. While most recent research has focused on how this ability develops during reinforcement learning, we show that it actually begins to emerge much earlier - during the model's pre-training. To study this, we introduce deliberate errors into chains-of-thought and test whether the model can still arrive at the correct answer by recognizing and correcting these mistakes. By tracking performance across different stages of pre-training, we observe that this self-correcting ability appears early and improves steadily over time. For instance, an OLMo2-7B model pre-trained on 4 trillion tokens displays self-correction on our six self-reflection tasks.

cs.CL↗

arXiv2025

Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control

We introduce Cosmos-Transfer, a conditional world generation model that can generate world simulations based on multiple spatial control inputs of various modalities such as segmentation, depth, and edge. In the design, the spatial conditional scheme is adaptive and customizable. It allows weighting different conditional inputs differently at different spatial locations. This enables highly controllable world generation and finds use in various world-to-world transfer use cases, including Sim2Real. We conduct extensive evaluations to analyze the proposed model and demonstrate its applications for Physical AI, including robotics Sim2Real and autonomous vehicle data enrichment. We further demonstrate an inference scaling strategy to achieve real-time world generation with an NVIDIA GB200 NVL72 rack. To help accelerate research development in the field, we open-source our models and code at https://github.com/nvidia-cosmos/cosmos-transfer1.

cs.CV↗

arXiv2025

GR00T N1: An Open Foundation Model for Generalist Humanoid Robots

General-purpose robots need a versatile body and an intelligent mind. Recent advancements in humanoid robots have shown great promise as a hardware platform for building generalist autonomy in the human world. A robot foundation model, trained on massive and diverse data sources, is essential for enabling the robots to reason about novel situations, robustly handle real-world variability, and rapidly learn new tasks. To this end, we introduce GR00T N1, an open foundation model for humanoid robots. GR00T N1 is a Vision-Language-Action (VLA) model with a dual-system architecture. The vision-language module (System 2) interprets the environment through vision and language instructions. The subsequent diffusion transformer module (System 1) generates fluid motor actions in real time. Both modules are tightly coupled and jointly trained end-to-end. We train GR00T N1 with a heterogeneous mixture of real-robot trajectories, human videos, and synthetically generated datasets. We show that our generalist robot model GR00T N1 outperforms the state-of-the-art imitation learning baselines on standard simulation benchmarks across multiple robot embodiments. Furthermore, we deploy our model on the Fourier GR-1 humanoid robot for language-conditioned bimanual manipulation tasks, achieving strong performance with high data efficiency.

cs.RO↗

arXiv2025

Every Sample Matters: Leveraging Mixture-of-Experts and High-Quality Data for Efficient and Accurate Code LLM

Recent advancements in code large language models (LLMs) have demonstrated remarkable capabilities in code generation and understanding. It is still challenging to build a code LLM with comprehensive performance yet ultimate efficiency. Many attempts have been released in the open source community to break the trade-off between performance and efficiency, such as the Qwen Coder series and the DeepSeek Coder series. This paper introduces yet another attempt in this area, namely Ling-Coder-Lite. We leverage the efficient Mixture-of-Experts (MoE) architecture along with a set of high-quality data curation methods (especially those based on program analytics) to build an efficient yet powerful code LLM. Ling-Coder-Lite exhibits on-par performance on 12 representative coding benchmarks compared to state-of-the-art models of similar size, such as Qwen2.5-Coder-7B and DeepSeek-Coder-V2-Lite, while offering competitive latency and throughput. In practice, we achieve a 50\% reduction in deployment resources compared to the similar-sized dense model without performance loss. To facilitate further research and development in this area, we open-source our models as well as a substantial portion of high-quality data for the annealing and post-training stages. The models and data can be accessed at~\url{https://huggingface.co/inclusionAI/Ling-Coder-lite}.

cs.LG↗

arXiv2025

Measurement of the Branching Fraction of $Λ_c^+ \to p K_S^0 π^0$ at Belle

We report a precise measurement of the ratio of branching fractions $\mathcal{B}(Λ_c^+\to p K_S^0 π^0)/\mathcal{B}(Λ_c^+\to p K^- π^+)$ using 980 fb$^{-1}$ of $e^+e^-$ data from the Belle experiment. We obtain a value of $\mathcal{B}(Λ_c^+\to p K_S^0 π^0)/\mathcal{B}(Λ_c^+\to p K^- π^+)=0.339\pm 0.002\pm 0.009$, where the first and second uncertainties are statistical and systematic, respectively. This Belle result is consistent with the previous measurement from the CLEO experiment but has a fivefold improvement in precision. By combining our result with the world average $\mathcal{B}(Λ_c^+\to p K^- π^+)$, we obtain the absolute branching fraction $\mathcal{B}(Λ_c^+\to p K_S^0 π^0)=(2.12\pm 0.01\pm 0.05 \pm 0.10)\%$, where the uncertainties are statistical, systematic, and the uncertainty in the absolute branching fraction scale $\mathcal{B}(Λ_c^+\to p K^- π^+)$, respectively. This measurement can shed light on hadronic decay mechanisms in charmed baryon decays.

hep-ex↗

arXiv2025

Detection of RS Oph with LST-1 and modelling of its HE/VHE gamma-ray emission

The recurrent nova RS Ophiuchi (RS Oph) underwent a thermonuclear eruption in August 2021. In this event, RS Oph was detected by the High Energy Stereoscopic System (H.E.S.S.), the Major Atmospheric Gamma Imaging Cherenkov (MAGIC), and the first Large-Sized Telescope (LST-1) of the future Cherenkov Telescope Array Observatory (CTAO) at very-high gamma-ray energies above 100 GeV. This means that novae are a new class of very-high-energy (VHE) gamma-ray emitters. We report the analysis of the RS Oph observations with LST-1. We constrain the particle population that causes the observed emission in hadronic and leptonic scenarios. Additionally, we study the prospects of detecting further novae using LST-1 and the upcoming LST array of CTAO-North. We conducted target-of-opportunity observations with LST-1 from the first day of this nova event. The data were analysed in the framework of cta-lstchain and Gammapy, the official CTAO-LST reconstruction and analysis packages. One-zone hadronic and leptonic models were considered to model the gamma-ray emission of RS Oph using the spectral information from Fermi-LAT and LST-1, together with public data from the MAGIC and H.E.S.S. telescopes. RS Oph was detected at $6.6σ$ with LST-1 in the first 6.35 hours of observations following the eruption. The hadronic scenario is preferred over the leptonic scenario considering a proton energy spectrum with a power-law model with an exponential cutoff whose position increases from $(0.26\pm 0.08)$ TeV on day 1 up to $(1.6\pm 0.6)$ TeV on day 4 after the eruption. The deep sensitivity and low energy threshold of the LST-1/LST array will allow us to detect faint novae and increase their discovery rate.

astro-ph.HE↗

arXiv2025

Measurements of higher-order cumulants of multiplicity and net-electric charge distributions in inelastic proton-proton interactions by NA61/SHINE

This paper presents the energy dependence of multiplicity and net-electric charge fluctuations in p+p interactions at beam momenta 20, 31, 40, 80, and 158 GeV/c. Results are corrected for the experimental biases and quantified with the use of cumulants and factorial cumulants. Cumulant ratios are an essential tool in the search for the critical point of strongly interacting matter in heavy ion collisions. Measurements performed in p+p interactions provide a vital baseline estimation in these studies. The measured signals are compared with the string hadronic models EPOS1.99 and FTFP-BERT.

hep-ex↗

arXiv2025

Nonperturbative aspects of the electromagnetic pion form factor at high energies

The structure of hadronic form factors at high energies and their deviations from perturbative quantum chromodynamics provide insight on nonperturbative dynamics. Using an approach that is consistent with dispersion relations, we construct a model that simultaneously accounts for the pion wave function, gluonic exchanges, and quark Reggeization. In particular, we find that quark Reggeization can be investigated at high energies by studying scaling violation of the form factor.

hep-ph↗

arXiv2025

Unveiling the Infrared Excess of SIPS J2045-6332: Evidence for a Young Stellar Object with Potential Low-Mass Companion

The Disk Detective project, a citizen science initiative, aims to identify circumstellar discs around stars by detecting objects with infrared (IR) excess using data from the Wide-field Infrared Survey Explorer (WISE). In this study, we investigate SIPS J2045-6332, a potential brown dwarf with significant IR excess in WISE and 2MASS bands, initially identified by project volunteers. Despite early indicators of a circumstellar disc, discrepancies between observed brightness and expected Spectral Energy Distribution (SED) models suggested unusual properties. To explore potential explanations, we created SED templates for spectral types M9 to L4 and compared them with SIPS J2045-6332's photometric data, revealing an excess brightness that points to either an unresolved low-mass companion or a young, inflated primary star. Further analysis of infrared spectral features and surface gravity indicators supports a youthful classification, estimating the object's age at 26-200 million years. Observations also suggest the presence of a mid L-type companion at a projected distance of 6.7 AU. This study highlights SIPS J2045-6332 as an intriguing system with unique IR characteristics and recommends follow-up observations with high-resolution telescopes to confirm the companion hypothesis and further characterize the system.

astro-ph.SR↗

arXiv2025

Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs

We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of 200K tokens to better support multilingual applications, as well as group query attention for more efficient long-sequence generation. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model. Its novel modality extension approach leverages LoRA adapters and modality-specific routers to allow multiple inference modes combining various modalities without interference. For example, it now ranks first in the OpenASR leaderboard to date, although the LoRA component of the speech/audio modality has just 460 million parameters. Phi-4-Multimodal supports scenarios involving (vision + language), (vision + speech), and (speech/audio) inputs, outperforming larger vision-language and speech-language models on a wide range of tasks. Additionally, we experiment to further train Phi-4-Mini to enhance its reasoning capabilities. Despite its compact 3.8-billion-parameter size, this experimental version achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.

cs.CL↗

arXiv2025

Measurement of the inclusive branching fractions for $B_s^0$ decays into $D$ mesons via hadronic tagging

We report measurements of the absolute branching fractions $\mathcal{B}(B_s^0 \to D_s^{\pm} X)$, $\mathcal{B}(B_s^0 \to D^0/\bar{D}^0 X)$, and $\mathcal{B}(B_s^0 \to D^{\pm} X)$, where the latter is measured for the first time. The results are based on a 121.4\,fb$^{-1}$ data sample collected at the $Υ(10860)$ resonance by the Belle detector at the KEKB asymmetric-energy $e^+ e^-$ collider. We reconstruct one $B_s^0$ meson in $e^+e^- \to Υ(10860) \to B_s^{*} \bar{B}_s^{*}$ events and measure yields of $D_s^+$, $D^0$, and $D^+$ mesons in the rest of the event. We obtain $\mathcal{B}(B_s^0 \to D_s^{\pm} X) = (68.6 \pm 7.2 \pm 4.0)\%$, $\mathcal{B}(B_s^0 \to D^0/\bar{D}^0 X) = (21.5 \pm 6.1 \pm 1.8)\%$, and $\mathcal{B}(B_s^0 \to D^{\pm} X) = (12.6 \pm 4.6 \pm 1.3)\%$, where the first uncertainty is statistical and the second is systematic. Averaging with previous Belle measurements gives $\mathcal{B}(B_s^0 \to D_s^{\pm} X) = (63.4 \pm 4.5 \pm 2.2)\%$ and $\mathcal{B}(B_s^0 \to D^0/\bar{D}^0 X) = (23.9 \pm 4.1 \pm 1.8)\%$. For the $B_s^0$ production fraction at the $Υ(10860)$, we find $f_s = (21.4^{+1.5}_{-1.7})\%$.

hep-ex↗

arXiv2025

Competitive Programming with Large Reasoning Models

We show that reinforcement learning applied to large language models (LLMs) significantly boosts performance on complex coding and reasoning tasks. Additionally, we compare two general-purpose reasoning models - OpenAI o1 and an early checkpoint of o3 - with a domain-specific system, o1-ioi, which uses hand-engineered inference strategies designed for competing in the 2024 International Olympiad in Informatics (IOI). We competed live at IOI 2024 with o1-ioi and, using hand-crafted test-time strategies, placed in the 49th percentile. Under relaxed competition constraints, o1-ioi achieved a gold medal. However, when evaluating later models such as o3, we find that o3 achieves gold without hand-crafted domain-specific strategies or relaxed constraints. Our findings show that although specialized pipelines such as o1-ioi yield solid improvements, the scaled-up, general-purpose o3 model surpasses those results without relying on hand-crafted inference heuristics. Notably, o3 achieves a gold medal at the 2024 IOI and obtains a Codeforces rating on par with elite human competitors. Overall, these results indicate that scaling general-purpose reinforcement learning, rather than relying on domain-specific techniques, offers a robust path toward state-of-the-art AI in reasoning domains, such as competitive programming.

cs.LG↗