Search

We investigate the low-filling spin-1 lattice gas in the intermediate interaction regime, in which the atom-atom interaction allows the decomposition of the system into the coupled spin and charge sectors, with lower energetical detuning between the two sectors than in the strong interaction regime. The low-lying eigenstates are grouped into different manifolds due to the decomposition, and are endowed with the emergent spatial inversion symmetry separately in the spin and charge sectors, which induces hidden correlations and affects the spin distribution of the system. The lowered energetical detuning between the two sectors activates the inter-sector coupling, and overlaps different manifolds in the eigenenergy spectrum, which leads to the crossings of eigenstates from different manifolds. The inter-sector coupling between the spin and charges is then witnessed by the the inter-manifold avoided crossings, which takes place between accidentally degenerate eigenstates of the same symmetry parity. Our work reveals the enhanced coupling effects between the spin and charge dopants of the spinor lattice gas in the intermediate interaction regime., Comment: 8 pages, 6 figures

We perform a simultaneous fit to BESIII data on $e^+e^-\to \gamma (D^0\bar{D}^{*0}\pi^0/J/\psi\pi^+\pi^-)$ and LHCb data on $B^+\to K^+(J/\psi\pi^+\pi^-)$ to precisely determine the properties of the $X(3872)$, with full consideration of three-body effects from $D^*\to D\pi$ decay, respecting both analyticity and unitarity. The $X(3872)$ is determined to be a bound state with a significance of $6\sigma$, and its pole is located at $E_X{=}\left(-53^{+~9}_{-24}-\,i\,34^{+~2}_{-12}\right) \rm keV$, relative to the nominal $D^0\bar{D}^{*0}$ threshold. Moreover, we confirm the presence of an isovector partner state, $W_{c1}$. It is found as a virtual state at $(1.6^{+0.7}_{-0.9}+ i\, 1.4^{+0.3}_{-0.6})\ \rm MeV$ relative to the $D^+ D^{*-}$ threshold on an unphysical Riemann sheet, strongly supporting a molecular nature of both $X(3872)$ and $W_{c1}$. As a highly non-trivial prediction we show that the $W_{c1}$ leads to nontrivial lineshapes in $B^0\to K^0X(3872)\to K^0 D^0\bar D^0\pi^0$ and $K^0J/\psi\pi^+\pi^-$ -- thus the scheme presented here can be tested further by improved measurements., Comment: 11 pages, 6 figures and 1 table

Preference optimization for diffusion models aims to align them with human preferences for images. Previous methods typically leverage Vision-Language Models (VLMs) as pixel-level reward models to approximate human preferences. However, when used for step-level preference optimization, these models face challenges in handling noisy images of different timesteps and require complex transformations into pixel space. In this work, we demonstrate that diffusion models are inherently well-suited for step-level reward modeling in the latent space, as they can naturally extract features from noisy latent images. Accordingly, we propose the Latent Reward Model (LRM), which repurposes components of diffusion models to predict preferences of latent images at various timesteps. Building on LRM, we introduce Latent Preference Optimization (LPO), a method designed for step-level preference optimization directly in the latent space. Experimental results indicate that LPO not only significantly enhances performance in aligning diffusion models with general, aesthetic, and text-image alignment preferences, but also achieves 2.5-28$\times$ training speedup compared to existing preference optimization methods. Our code will be available at https://github.com/casiatao/LPO., Comment: 20 pages, 14 tables, 15 figures

Current autonomous vehicles operate primarily within limited regions, but there is increasing demand for broader applications. However, as models scale, their limited capacity becomes a significant challenge for adapting to novel scenarios. It is increasingly difficult to improve models for new situations using a single monolithic model. To address this issue, we introduce the concept of dynamically enhancing a basic driving planner with local driving data, without permanently modifying the planner itself. This approach, termed the Dynamically Local-Enhancement (DLE) Planner, aims to improve the scalability of autonomous driving systems without significantly expanding the planner's size. Our approach introduces a position-varying Markov Decision Process formulation coupled with a graph neural network that extracts region-specific driving features from local observation data. The learned features describe the local behavior of the surrounding objects, which is then leveraged to enhance a basic reinforcement learning-based policy. We evaluated our approach in multiple scenarios and compared it with a one-for-all driving model. The results show that our method outperforms the baseline policy in both safety (collision rate) and average reward, while maintaining a lighter scale. This approach has the potential to benefit large-scale autonomous vehicles without the need for largely expanding on-device driving models.

Gradient-based methods are well-suited for derivative-free optimization (DFO), where finite-difference (FD) estimates are commonly used as gradient surrogates. Traditional stochastic approximation methods, such as Kiefer-Wolfowitz (KW) and simultaneous perturbation stochastic approximation (SPSA), typically utilize only two samples per iteration, resulting in imprecise gradient estimates and necessitating diminishing step sizes for convergence. In this paper, we first explore an efficient FD estimate, referred to as correlation-induced FD estimate, which is a batch-based estimate. Then, we propose an adaptive sampling strategy that dynamically determines the batch size at each iteration. By combining these two components, we develop an algorithm designed to enhance DFO in terms of both gradient estimation efficiency and sample efficiency. Furthermore, we establish the consistency of our proposed algorithm and demonstrate that, despite using a batch of samples per iteration, it achieves the same convergence rate as the KW and SPSA methods. Additionally, we propose a novel stochastic line search technique to adaptively tune the step size in practice. Finally, comprehensive numerical experiments confirm the superior empirical performance of the proposed algorithm.

We report the phenomena of electromagnetically induced transparency (EIT) and electromagnetically induced absorption (EIA) using two identical beams in rubidium atomic vapor. The {\Lambda}-type EIT configuration is employed to examine the EIT spectrum for the D1 line in 87Rb F=2 characteristics16 by varying parameters such as frequency detuning, Iprobe/Ipump, the total power of probe and pump beam. Notably, the pump beam is also investigated in this process, which has not been previously studied. We study the effect of of the phase between the two applied fields and find that EIA and EIT can transform into each other by adjusting the relative phase. These finding may have applications in light drag or storage, optical switching, and sensing.

The exploration of the surrounding world and the universe is an important theme in the legacy of humankind. The detection of gravitational waves is adding a new dimension to this grand effort. What are the fundamental physical laws governing the dynamics of the universe? What is the fundamental composition of the universe? How has the universe evolved in the past and how will it evolve in the future? These are the basic questions that press for answers. The space-based gravitational wave detector TianQin will tune in to gravitational waves in the millihertz frequency range ($10^{-4} \sim 1$ Hz, to be specific), opening a new gravitational wave spectrum window to explore many of the previously hidden sectors of the universe. TianQin will discover many astrophysical systems, populating the universe at different redshifts: some will be of new types that have never been detected before, some will have very high signal-to-noise ratios, and some will have very high parameter estimation precision. The plethora of information collected will bring us to new fronts on which to search for the breaking points of general relativity, the possible violation of established physical laws, the signature of possible new gravitational physics and new fundamental fields, and to improve our knowledge on the expansion history of the universe. In this white paper, we highlight the advances that TianQin can bring to fundamental physics and cosmology., Comment: 113 pages

Automated radiology report generation offers an effective solution to alleviate radiologists' workload. However, most existing methods focus primarily on single or fixed-view images to model current disease conditions, which limits diagnostic accuracy and overlooks disease progression. Although some approaches utilize longitudinal data to track disease progression, they still rely on single images to analyze current visits. To address these issues, we propose enhanced contrastive learning with Multi-view Longitudinal data to facilitate chest X-ray Report Generation, named MLRG. Specifically, we introduce a multi-view longitudinal contrastive learning method that integrates spatial information from current multi-view images and temporal information from longitudinal data. This method also utilizes the inherent spatiotemporal information of radiology reports to supervise the pre-training of visual and textual representations. Subsequently, we present a tokenized absence encoding technique to flexibly handle missing patient-specific prior knowledge, allowing the model to produce more accurate radiology reports based on available prior knowledge. Extensive experiments on MIMIC-CXR, MIMIC-ABN, and Two-view CXR datasets demonstrate that our MLRG outperforms recent state-of-the-art methods, achieving a 2.3% BLEU-4 improvement on MIMIC-CXR, a 5.5% F1 score improvement on MIMIC-ABN, and a 2.7% F1 RadGraph improvement on Two-view CXR., Comment: Accepted by CVPR 2025

The dominant fraction of the extragalactic $\gamma$-ray sources are blazars, active galactic nuclei with jets inclined at a small angle to the line of sight. Apart from blazars, a few dozen narrow-line Seyfert 1 galaxies (NLS1) and a number of radio galaxies are associated with $\gamma$-ray sources. The identification of $\gamma$-ray sources requires multiwavelength follow-up observations since several candidates could reside within the relatively large $\gamma$-ray localisation area. The $\gamma$-ray source 4FGL 0959.6+4606 was originally associated with a radio galaxy. However, follow-up multiwavelength work suggested a nearby NLS1 as the more probable origin of the $\gamma$-ray emission. We performed high-resolution very long baseline interferometry (VLBI) observation at 5 GHz of both proposed counterparts of 4FGL 0959.6+4606. We clearly detected the NLS1 source SDSS J095909.51+460014.3 with relativistically boosted jet emission. On the other hand, we did not detect milliarcsecond-scale compact emission in the radio galaxy 2MASX J09591976+4603515. Our VLBI imaging results suggest that the NLS1 object is the origin of the $\gamma$-ray emission in 4FGL 0959.6+4606., Comment: 11 pages, 1 figure, accepted for publication in Universe

Vision-language models (VLMs) have made significant progress in image classification by training with large-scale paired image-text data. Their performances largely depend on the prompt quality. While recent methods show that visual descriptions generated by large language models (LLMs) enhance the generalization of VLMs, class-specific prompts may be inaccurate or lack discrimination due to the hallucination in LLMs. In this paper, we aim to find visually discriminative prompts for fine-grained categories with minimal supervision and no human-in-the-loop. An evolution-based algorithm is proposed to progressively optimize language prompts from task-specific templates to class-specific descriptions. Unlike optimizing templates, the search space shows an explosion in class-specific candidate prompts. This increases prompt generation costs, iterative times, and the overfitting problem. To this end, we first introduce several simple yet effective edit-based and evolution-based operations to generate diverse candidate prompts by one-time query of LLMs. Then, two sampling strategies are proposed to find a better initial search point and reduce traversed categories, saving iteration costs. Moreover, we apply a novel fitness score with entropy constraints to mitigate overfitting. In a challenging one-shot image classification setting, our method outperforms existing textual prompt-based methods and improves LLM-generated description methods across 13 datasets. Meanwhile, we demonstrate that our optimal prompts improve adapter-based methods and transfer effectively across different backbones., Comment: Accepted to the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025

IceCube has observed a diffuse astrophysical neutrino flux over the energy region from a few TeV to a few PeV. At PeV energies, the spectral shape is not yet well measured due to the low statistics of the data. This analysis probes the gap between 1 PeV and 10 PeV by using high-energy downgoing muon neutrinos. To reject the large atmospheric muon background, two complementary techniques are combined. The first technique selects events with high stochasticity to reject atmospheric muon bundles whose stochastic energy losses are smoothed due to high muon multiplicity. The second technique vetoes atmospheric muons with the IceTop surface array. Using 9 years of data, we found two neutrino candidate events in the signal region, consistent with expectation from background, each with relatively high signal probabilities. A joint maximum likelihood estimation is performed using this sample and an independent 9.5-year sample of tracks to measure the neutrino spectrum. A likelihood ratio test is done to compare the single power-law (SPL) vs. SPL+cutoff hypothesis; the SPL+cutoff model is not significantly better than the SPL. High-energy astrophysical objects from four source catalogs are also checked around the direction of the two events. No significant coincidence was found.

The validity range of the widely used traditional effective range expansion can be severely limited by the presence of a left-hand cut near the two-particle threshold. Such a left-hand cut emerges in two-particle scattering processes involving either a light particle exchange in the $t$-channel or a particle exchange with a mass slightly heavier than the mass difference of the two particles in the $u$-channel, which occurs in a wide range of physical systems. We propose a new parameterization for the low-energy scattering amplitude that incorporates these left-hand cuts arising from particle exchange diagrams. This parameterization extends the convergence radius of the effective range expansion beyond the branch point of the left-hand cut and is applicable to a broad range of systems. The parameterization enables the extraction of coupling strengths between the exchange particle and the scattering particles, and reveals amplitude zeros resulting from the interplay between short- and long-range interactions. We demonstrate the effectiveness of this new parameterization through its application to $DD^*$ scattering with meson masses obtained in a lattice QCD calculation., Comment: Proceedings of the 11th International Workshop on Chiral Dynamics (CD2024)

In practice, physical spatiotemporal forecasting can suffer from data scarcity, because collecting large-scale data is non-trivial, especially for extreme events. Hence, we propose \method{}, a novel probabilistic framework to realize iterative self-training with new self-ensemble strategies, achieving better physical consistency and generalization on extreme events. Following any base forecasting model, we can encode its deterministic outputs into a latent space and retrieve multiple codebook entries to generate probabilistic outputs. Then BeamVQ extends the beam search from discrete spaces to the continuous state spaces in this field. We can further employ domain-specific metrics (e.g., Critical Success Index for extreme events) to filter out the top-k candidates and develop the new self-ensemble strategy by combining the high-quality candidates. The self-ensemble can not only improve the inference quality and robustness but also iteratively augment the training datasets during continuous self-training. Consequently, BeamVQ realizes the exploration of rare but critical phenomena beyond the original dataset. Comprehensive experiments on different benchmarks and backbones show that BeamVQ consistently reduces forecasting MSE (up to 39%), enhancing extreme events detection and proving its effectiveness in handling data scarcity.

We calculate the nucleon mass in a manifestly relativistic baryon chiral perturbation theory up to the leading two-loop order. Through dimensional counting analysis, we perform the chiral expansion and verify the validity of the extended-on-mass-shell scheme at the two-loop level. As a result, we obtain the complete chiral representation of the nucleon mass up to $\mathcal{O}(p^5)$, which preserves the original analytic properties and satisfies the correct power counting. The obtained chiral result is well-suited for chiral extrapolation and provides an excellent description of lattice QCD data across a broad range of pion masses. We find that the $\mathcal{O}(p^5)$ contribution is small, approximately $10$ MeV, and varies only mildly with increasing pion mass, demonstrating good convergence of the nucleon mass up to pion masses of about 350 MeV at two-loop order., Comment: 37 pages, 9 figures, 2 tables, comments are welcome!

The integration of quantum communication protocols over Ethernet networks is proposed, showing the potential of combining classical and quantum technologies for efficient, scalable quantum networking. By leveraging the inherent strengths of Ethernet, such as addressing, MAC layer functionality, and scalability; we propose a practical framework to support the rigorous requirements of quantum communication. Some novel protocols given in this study enable reliable end-to-end quantum entanglement over Ethernet, ensuring the adaptability needed for implementing a stable quantum internet. Detailed time-delay analyses confirm that our protocols offer superior performance compared to existing methods, with total time delay kept within the decoherence threshold of qubits. These results suggest that our approach is well-suited for deployment in realistic environments, meeting both the immediate needs of quantum networking and laying the groundwork for future advances in data exchange and quantum computational capabilities., Comment: Accepted in QCNC 2025 Nara, Japan

Significant progress has been achieved on the improvement and downstream usages of the Contrastive Language-Image Pre-training (CLIP) vision-language model, while less attention is paid to the interpretation of CLIP. We propose a Gradient-based visual and textual Explanation method for CLIP (Grad-ECLIP), which interprets the matching result of CLIP for specific input image-text pair. By decomposing the architecture of the encoder and discovering the relationship between the matching similarity and intermediate spatial features, Grad-ECLIP produces effective heat maps that show the influence of image regions or words on the CLIP results. Different from the previous Transformer interpretation methods that focus on the utilization of self-attention maps, which are typically extremely sparse in CLIP, we produce high-quality visual explanations by applying channel and spatial weights on token features. Qualitative and quantitative evaluations verify the effectiveness and superiority of Grad-ECLIP compared with the state-of-the-art methods. Furthermore, a series of analysis are conducted based on our visual and textual explanation results, from which we explore the working mechanism of image-text matching, the strengths and limitations in attribution identification of CLIP, and the relationship between the concreteness/abstractness of a word and its usage in CLIP. Finally, based on the ability of explanation map that indicates text-specific saliency region of input image, we also propose an application with Grad-ECLIP, which is adopted to boost the fine-grained alignment in the CLIP fine-tuning. The code of Grad-ECLIP is available here: https://github.com/Cyang-Zhao/Grad-Eclip.

We identified blue straggler stars (BSSs) in 53 open clusters utilizing data from Gaia DR3. Most of these clusters are situated in the outer regions of the Galactic disc, encompassing structures such as the warp and the Outer arm. We analyzed their astrometric parameters and determined that 48 of them demonstrate high reliability in radial density profile. Furthermore, through manual isochrone fitting and visual inspection, we confirmed 119 BSS candidates and identified 328 additional possible candidates within these clusters. Our results contribute to a 46% increase in the sample size of BSSs in open clusters for regions of the Galactic disc where Rgc > 12 kpc. We observed that the new samples are fainter compared to those identified in the past. Additionally, we investigated the maximum fractional mass excess (Me) of the BSSs in open clusters, including previously published BSS samples. Our findings indicate a strong correlation between the capability to produce highest-Me BSSs and the mass of their host clusters. This observation appears to reinforce a fundamental principle whereby an increase in the mass of a star cluster correlates with a higher likelihood of stellar mergers. In contrast, we observe minimal correlation between maximum-Me and the cluster age. Among clusters containing BSSs, younger clusters (0.5 to 1 Gyr) display a scarcity of high-Me BSSs. This scarcity may be attributed to the absence of more massive clusters within this age range., Comment: 18 pages, AJ accepted

Large Language Models (LLMs) possess human-level cognitive and decision-making capabilities, making them a key technology for 6G. However, applying LLMs to the communication domain faces three major challenges: 1) Inadequate communication data; 2) Restricted input modalities; and 3) Difficulty in knowledge retrieval. To overcome these issues, we propose CommGPT, a multimodal foundation model designed specifically for communications. First, we create high-quality pretraining and fine-tuning datasets tailored in communication, enabling the LLM to engage in further pretraining and fine-tuning with communication concepts and knowledge. Then, we design a multimodal encoder to understand and process information from various input modalities. Next, we construct a Graph and Retrieval-Augmented Generation (GRG) framework, efficiently coupling Knowledge Graph (KG) with Retrieval-Augmented Generation (RAG) for multi-scale learning. Finally, we demonstrate the feasibility and effectiveness of the CommGPT through experimental validation.

This study presents an energy-dependent analysis of seasonal variations in the atmospheric muon neutrino spectrum, using 11.3 years of data from the IceCube Neutrino Observatory. By leveraging a novel spectral unfolding method, we explore the energy range from 125 GeV to 10 TeV for zenith angles between 90{\deg} to 110{\deg}, corresponding to the Antarctic atmosphere. Our findings reveal that the seasonal variation amplitude decreases with energy reaching ($-4.6 \pm 1.1$)\% during Austral winter and increases ($+3.9 \pm 1.2$)\% during Austral summer relative to the annual average at 10TeV. While the unfolded flux exceeds the model predictions by up to 30\%, the differential measurement of seasonal variations remains unaffected. The measured seasonal variations of the muon neutrino spectrum are consistent with theoretical predictions using the MCEq code and the NRLMSISE-00 atmospheric model.

The theoretical maximum critical temperature ($T_c$) for conventional superconductors at ambient pressure remains a fundamental question in condensed matter physics. Through analysis of electron-phonon calculations for over 20,000 metals, we critically examine this question. We find that while hydride metals can exhibit maximum phonon frequencies of more than 5000 K, the crucial logarithmic average frequency $\omega_\text{log}$ rarely exceeds 1800 K. Our data reveals an inherent trade-off between $\omega_\text{log}$ and the electron-phonon coupling constant $\lambda$, suggesting that the optimal Eliashberg function that maximizes $T_c$ is unphysical. Based on our calculations, we identify Li$_2$AgH$_6$ and its sibling Li$_2$AuH$_6$ as theoretical materials that likely approach the practical limit for conventional superconductivity at ambient pressure. Analysis of thermodynamic stability indicates that compounds with higher predicted $T_c$ values are increasingly unstable, making their synthesis challenging. While fundamental physical laws do not strictly limit $T_c$ to low-temperatures, our analysis suggests that achieving room-temperature conventional superconductivity at ambient pressure is extremely unlikely.

Interfacial superconductivity discovered in bilayer nickelate heterostructures opens new avenues for effective tunability and control. Here, we report the superconductor-insulator transition in La$_{2.85}$Pr$_{0.15}$Ni$_2$O$_7$/SrLaAlO$_4$ interfaces, with corresponding electronic structures revealed using synchrotron low-temperature X-ray absorption spectroscopy (XAS) and X-ray linear dichroism (XLD). In addition to the valence states of each element within the films determined by XAS, XLD at the O K edge pre-peak reveals hybridizations involving both 2p$_{z}$ and 2p$_{x}$/2p$_{y}$ orbitals, which are concomitantly suppressed upon oxygen loss. Intriguingly, such oxygen loss that eliminates superconductivity, induces a qualitative XLD change at the Ni L$_{2}$ edge, implying significant transformations among 3d$_{x^{2}}$ $_{-y^{2}}$ and 3d$_{z^{2}}$ energy bands near the Fermi level. Our findings establish an element- and orbital-resolved electronic framework for the origin of superconductivity., Comment: under review

Controlling quantum systems with time-dependent fields opens avenues for engineering novel states of matter and exploring non-equilibrium phenomena. Landau levels in two-dimensional electron gases (2DEGs), with their discrete energy spectrum and characteristic cyclotron dynamics, provide an important platform for realizing and studying such driven quantum systems. While exact solutions for driven Landau levels exist, they have been limited to specific gauges or representations. In this work, we present an algebraic, gauge- and representation-independent exact solution for driven Landau levels in 2DEGs subject to arbitrary time-dependent electromagnetic fields. Our approach, based on a time-dependent unitary transformation via the displacement operator, provides clear physical insights into the driven quantum dynamics. We apply this method to derive the exact Floquet states and quasienergies for periodically driven Landau levels, and we extend our analysis to the resonant driving regime, where the Floquet picture breaks down and the electron wavefunction exhibits unbounded spatial spreading. Furthermore, we calculate the instantaneous energy absorption rate, revealing distinct absorption behaviors between coherent states and Fock or thermal states, stemming from quantum interference effects.

Laser excitation has emerged as an effective tool for probing microscopic interactions and manipulating phases of matter. Among charge density wave (CDW) materials, 1T-TaS2 has garnered significant attention due to its diverse stacking orders and photoexcited responses. However, the mechanisms driving transitions among different stacking orders and the microscopic out-of-equilibrium dynamics remain unclear. We elucidate that photoexcitation can introduce interlayer stacking order transitions facilitated by laser-induced intralayer reconstruction in 1T-TaS2. Importantly, our finding reveals a novel pathway to introduce different phases through laser excitations, apparently distinct from thermally-induced phase transitions via interlayer sliding. In particular, photoexcitation is able to considerably change potential energy surfaces and evoke collective lattice dynamics. Consequently, the laser-induced intralayer reconstruction plays a crucial role in interlayer stacking-order transition, offering a new method to create exotic stackings and quantum phases. The exploration opens up great opportunities for manipulating CDW phases and electronic properties on the femtosecond timescale., Comment: 10 pages, 4 figures

Users and creators are two crucial components of recommender systems. Typical recommender systems focus on the user side, providing the most suitable items based on each user's request. In such scenarios, a few items receive a majority of exposures, while many items receive very few. This imbalance leads to poorer experiences and decreased activity among the creators receiving less feedback, harming the recommender system in the long term. To this end, we develop a creator-side recommender system, called DualRec, to answer the following question: how to find the most suitable users for each item to enhance the creators' experience? We show that typical user-side recommendation algorithms, such as retrieval and ranking algorithms, can be adapted into the creator-side versions with just a few modifications. This greatly simplifies algorithm design in DualRec. Moreover, we discuss a unique challenge in DualRec: the user availability issue, which is not present in user-side recommender systems. To tackle this issue, we incorporate a user availability calculation (UAC) module to effectively enhance DualRec's performance. DualRec has already been implemented in Kwai, a short video recommendation system with over 100 millions user and over 10 million creators, significantly improving the experience for creators., Comment: 9 pages and 9 figures

The discovery of superconductivity in bilayer nickelate thin films under ambient pressure provides an opportunity to directly investigate characteristic energy scales of the superconducting state from electronic structure. Here, we successfully probe the energy gap and dispersion renormalization in one unit-cell La$_{2.85}$Pr$_{0.15}$Ni$_2$O$_7$ films epitaxially grown on SrLaAlO$_4$ substrates, by developing an ultra-high vacuum quenching technique for in-situ angle-resolved photoemission spectroscopy measurements. The energy gap is observed on the underlying Fermi surface without showing a node along the Brillouin zone diagonal. This gap exhibits particle-hole symmetric behavior and persists to a temperature higher than the superconducting transition temperature, indicating the existence of a pseudogap. An abrupt band renormalization is observed with a dispersion anomaly at ~70 meV below Fermi level, pointing to another energy scale besides the energy gap. These observations provide initial information on fundamental properties of the superconducting state in bilayer nickelates under ambient pressure.

Rotating particles moving in fluids undergo a transverse migration via the inertia-induced Magnus effect. This phenomenon vanishes at colloidal scales because inertia is negligible and the fluid flow is time reversible. Yet, recent experiments discovered an inverse Magnus effect of colloids in polymeric and micellar solutions supposedly because their viscoelasticity breaks the time reversibility. Our study shows that classical viscoelastic features -- normal-stress differences and/or shear-thinning cannot explain this phenomenon. Instead, it originates from local polymer density inhomogeneities due to their stress-gradient-induced transport, a mechanism increasingly important at smaller scales -- indeed relevant to colloidal experiments. Incorporating this mechanism into our model leads to quantitative agreement with the experiments without fitting parameters. Our work provides new insights into colloidal motion in complex fluids with microstructural inhomogeneities, offers a simple mechanistic theory for predicting the resulting migration, and underscores the necessity of assimilating these findings in future designs of micro-machinery including swimmers, actuators, rheometers, and so on.

The kagome metals $A$V$_3$Sb$_5$ ($A=$ K, Cs, Rb) have become a fascinating materials platform following the discovery of many novel quantum states due to the interplay between electronic correlation, topology, and geometry. Understanding their physical origin requires constructing effective theories that capture the low-energy electronic structure and electronic interactions. While the band structure calculated by density functional theory (DFT) broadly agrees with experiments in the unbroken symmetry phase, the multiorbital nature challenges a proper understanding of the band structure and its description by tight-binding models. Here, we point out the unusual and puzzling properties of the DFT electronic structure, including the sublattice type of the van Hove singularities, the geometric shape of the Fermi surface, and the orbital content of the low-energy band dispersion, which cannot be described by the commonly used one-orbital or multiorbital kagome tight-binding models. We address these fundamental puzzles and develop an extended Slater-Koster formalism that can successfully resolve these issues. We discover the important role of site-symmetry and interorbital hopping structure and provide a concrete multiorbital tight-binding model description of the electronic structure for $A$V$_3$Sb$_5$ and the family of ``135'' compounds with other transition metals. This is a crucial step toward studying the effects of electron-electron interactions for the correlated and topological states in kagome metals and superconductors., Comment: 14 pages, 9 figures

In general, humans would grasp an object differently for different tasks, e.g., "grasping the handle of a knife to cut" vs. "grasping the blade to hand over". In the field of robotic grasp pose detection research, some existing works consider this task-oriented grasping and made some progress, but they are generally constrained by low-DoF gripper type or non-cluttered setting, which is not applicable for human assistance in real life. With an aim to get more general and practical grasp models, in this paper, we investigate the problem named Task-Oriented 6-DoF Grasp Pose Detection in Clutters (TO6DGC), which extends the task-oriented problem to a more general 6-DOF Grasp Pose Detection in Cluttered (multi-object) scenario. To this end, we construct a large-scale 6-DoF task-oriented grasping dataset, 6-DoF Task Grasp (6DTG), which features 4391 cluttered scenes with over 2 million 6-DoF grasp poses. Each grasp is annotated with a specific task, involving 6 tasks and 198 objects in total. Moreover, we propose One-Stage TaskGrasp (OSTG), a strong baseline to address the TO6DGC problem. Our OSTG adopts a task-oriented point selection strategy to detect where to grasp, and a task-oriented grasp generation module to decide how to grasp given a specific task. To evaluate the effectiveness of OSTG, extensive experiments are conducted on 6DTG. The results show that our method outperforms various baselines on multiple metrics. Real robot experiments also verify that our OSTG has a better perception of the task-oriented grasp points and 6-DoF grasp poses., Comment: Accepted to ICRA 2025

The non-detection of periodicity related to rotation challenges the magnetar model for fast radio bursts (FRBs). Moreover, a bimodal distribution of the burst waiting times is widely observed in hyper-active FRBs, a significant deviation from the exponential distribution expected from stationary Poisson processes. By combining the epidemic-type aftershock sequence (ETAS) earthquake model and the rotating vector model (RVM) involving the rotation of the magnetar and orientations of the spin and magnetic axes, we find that starquake events modulated by the rotation of FRB-emitting magnetar can explain the bimodal distribution of FRB waiting times, as well as the non-detection of periodicity in active repeating FRBs. We analyze data from multiple FRB sources, demonstrating that differences in waiting time distributions and observed energies can be explained by varying parameters related to magnetar properties and starquake dynamics. Our results suggest that rotation-modulated starquakes on magnetars can possibly be a unified source for FRBs. Notably, we find that active repeaters tend to have small magnetic inclination angles in order to hide their periodicity. We also show that our model can reproduce the waiting time distribution of a pulsar phase of the galactic magnetar SGR J1935+2154 with a larger inclination angle than the active repeaters, which could explain the detection of spin period and the relatively low observed energy for FRBs from the magnetar. The spin periods of active repeaters are not well constrained, but most likely fall in the valley region between the two peaks of the waiting time distributions., Comment: 27 Pages, 18 Figures, 2 Tables. Submitted to ApJ

Rumours in online social media pose significant risks to modern society, motivating the need for better understanding of how they develop. We focus specifically on the interface between emotion and rumours in threaded discourses, building on the surprisingly sparse literature on the topic which has largely focused on emotions within the original rumour posts themselves, and largely overlooked the comparative differences between rumours and non-rumours. In this work, we provide a comprehensive analytical emotion framework, contrasting rumour and non-rumour cases using existing NLP datasets to further understand the emotion dynamics within rumours. Our framework reveals several findings: rumours exhibit more negative sentiment and emotions, including anger, fear and pessimism, while non-rumours evoke more positive emotions; emotions are contagious in online interactions, with rumours facilitate negative emotions and non-rumours foster positive emotions; and based on causal analysis, surprise acts as a bridge between rumours and other emotions, pessimism is driven by sadness and fear, optimism by joy and love., Comment: 11 pages, 10 figures

Multi-agent reinforcement learning (MARL) faces challenges in coordinating agents due to complex interdependencies within multi-agent systems. Most MARL algorithms use the simultaneous decision-making paradigm but ignore the action-level dependencies among agents, which reduces coordination efficiency. In contrast, the sequential decision-making paradigm provides finer-grained supervision for agent decision order, presenting the potential for handling dependencies via better decision order management. However, determining the optimal decision order remains a challenge. In this paper, we introduce Action Generation with Plackett-Luce Sampling (AGPS), a novel mechanism for agent decision order optimization. We model the order determination task as a Plackett-Luce sampling process to address issues such as ranking instability and vanishing gradient during the network training process. AGPS realizes credit-based decision order determination by establishing a bridge between the significance of agents' local observations and their decision credits, thus facilitating order optimization and dependency management. Integrating AGPS with the Multi-Agent Transformer, we propose the Prioritized Multi-Agent Transformer (PMAT), a sequential decision-making MARL algorithm with decision order optimization. Experiments on benchmarks including StarCraft II Multi-Agent Challenge, Google Research Football, and Multi-Agent MuJoCo show that PMAT outperforms state-of-the-art algorithms, greatly enhancing coordination efficiency., Comment: Accepted by AAMAS 2025

Semantic Communication (SemCom) is a promising new paradigm for next-generation communication systems, emphasizing the transmission of core information, particularly in environments characterized by uncertainty, noise, and bandwidth constraints. However, existing image SemCom systems face several challenges, such as inefficient knowledge base construction, insufficient semantic encoding, and lack of multi-user semantic sharing. To address these issues, we propose a Lightweight Vision Model-based Multi-user Semantic Communication System (LVM-MSC). First, we construct a Lightweight Knowledge Base (LKB) based on the fast Segment Anything Model (SAM). LKB incorporates the extensive image knowledge of the SAM model while significantly reducing the number of parameters through its convolutional architecture. Next, we design an Efficient Semantic Codec (ESC) based on the Masked AutoEncoder (MAE) architecture. ESC enhances semantic compression at both the pixel and semantic levels and implements lightweight semantic decoding tailored for user devices. Furthermore, we propose a Multi-user Semantic Sharing (MSS) transmission for the multi-user SemCom. By calculating the similarity of semantic information among different users in the sharing semantic space, we unify the transmissions of similar semantic information through broadcasting, further improving the transmission efficiency. Finally, simulation results demonstrate the feasibility and effectiveness of the proposed LVM-MSC system.

Multi-modal Large Language Models (MLLMs) are capable of precisely extracting high-level semantic information from multi-modal data, enabling multi-task understanding and generation. This capability facilitates more efficient and intelligent data transmission in semantic communications. In this paper, we design a tailored MLLM for semantic communication and propose an MLLM-based Multi-modal, Multi-task and Multi-user Semantic Communication (M4SC) system. First, we utilize the Kolmogorov-Arnold Network (KAN) to achieve multi-modal alignment in MLLMs, thereby enhancing the accuracy of semantics representation in the semantic space across different modalities. Next, we introduce a multi-task fine-tuning approach based on task instruction following, which leverages a unified task instruction template to describe various semantic communication tasks, improving the MLLM's ability to follow instructions across multiple tasks. Additionally, by designing a semantic sharing mechanism, we transmit the public and private semantic information of multiple users separately, thus increasing the efficiency of semantic communication. Finally, we employ a joint KAN-LLM-channel coding strategy to comprehensively enhance the performance of the semantic communication system in complex communication environments. Experimental results validate the effectiveness and robustness of the proposed M4SC in multi-modal, multi-task, and multi-user scenarios.

Evaluating large language models (LLMs) poses significant challenges, particularly due to issues of data contamination and the leakage of correct answers. To address these challenges, we introduce ThinkBench, a novel evaluation framework designed to evaluate LLMs' reasoning capability robustly. ThinkBench proposes a dynamic data generation method for constructing out-of-distribution (OOD) datasets and offers an OOD dataset that contains 2,912 samples drawn from reasoning tasks. ThinkBench unifies the evaluation of reasoning models and non-reasoning models. We evaluate 16 LLMs and 4 PRMs under identical experimental conditions and show that most of the LLMs' performance are far from robust and they face a certain level of data leakage. By dynamically generating OOD datasets, ThinkBench effectively provides a reliable evaluation of LLMs and reduces the impact of data contamination.

In recent years, wearable devices have revolutionized cardiac monitoring by enabling continuous, non-invasive ECG recording in real-world settings. Despite these advances, the accuracy of ECG parameter calculations (PR interval, QRS interval, QT interval, etc.) from wearables remains to be rigorously validated against conventional ECG machines and expert clinician assessments. In this large-scale, multicenter study, we evaluated FeatureDB, a novel algorithm for automated computation of ECG parameters from wearable single-lead signals Three diverse datasets were employed: the AHMU-FH dataset (n=88,874), the CSE dataset (n=106), and the HeartVoice-ECG-lite dataset (n=369) with annotations provided by two experienced cardiologists. FeatureDB demonstrates a statistically significant correlation with key parameters (PR interval, QRS duration, QT interval, and QTc) calculated by standard ECG machines and annotated by clinical doctors. Bland-Altman analysis confirms a high level of agreement.Moreover,FeatureDB exhibited robust diagnostic performance in detecting Long QT syndrome (LQT) and atrioventricular block interval abnormalities (AVBI),with excellent area under the ROC curve (LQT: 0.836, AVBI: 0.861),accuracy (LQT: 0.856, AVBI: 0.845),sensitivity (LQT: 0.815, AVBI: 0.877),and specificity (LQT: 0.856, AVBI: 0.845).This further validates its clinical reliability. These results validate the clinical applicability of FeatureDB for wearable ECG analysis and highlight its potential to bridge the gap between traditional diagnostic methods and emerging wearable technologies.Ultimately,this study supports integrating wearable ECG devices into large-scale cardiovascular disease management and early intervention strategies,and it highlights the potential of wearable ECG technologies to deliver accurate,clinically relevant cardiac monitoring while advancing broader applications in cardiovascular care., Comment: 37 pages, 8 figures, 6 tables

Inspired by the states $X(2370)$ and $X(2600)$ reported by the BESIII Collaboration, we systematically investigate the mass spectra of light compact tetraquark states with configurations $ud\bar{u}\bar{d}$, $us\bar{u}\bar{s}$, and $ss\bar{s}\bar{s}$ in the $J^{PC}=0^{-+}$ and $2^{-+}$ channels using the QCD sum rules approach. To effectively describe these tetraquark states, we construct appropriate interpolating tetraquark currents featuring three Lorentz indices while avoiding derivative operators. Through meticulous calculations of correlation functions up to dimension 10 condensates, we extract the mass spectra for both $0^{-+}$ and $2^{-+}$ states by employing the projection operator technique. Our results indicate that the masses of light tetraquarks span $1.5-2.5~\text{GeV}$ for $0^{-+}$ states and $2.4-2.7~\text{GeV}$ for $2^{-+}$ states. Notably, our analysis suggests that the $X(2370)$ state could be interpreted as a $0^{-+}$ $us\bar{u}\bar{s}$ or $ss\bar{s}\bar{s}$ tetraquark state, while the $X(2600)$ state is likely to be a $2^{-+}$ $us\bar{u}\bar{s}$ tetraquark. These intriguing findings warrant further detailed investigation in future studies to better understand the nature of these states.

Text-to-SQL models, which parse natural language (NL) questions to executable SQL queries, are increasingly adopted in real-world applications. However, deploying such models in the real world often requires adapting them to the highly specialized database schemas used in specific applications. We find that existing text-to-SQL models experience significant performance drops when applied to new schemas, primarily due to the lack of domain-specific data for fine-tuning. This data scarcity also limits the ability to effectively evaluate model performance in new domains. Continuously obtaining high-quality text-to-SQL data for evolving schemas is prohibitively expensive in real-world scenarios. To bridge this gap, we propose SQLsynth, a human-in-the-loop text-to-SQL data annotation system. SQLsynth streamlines the creation of high-quality text-to-SQL datasets through human-LLM collaboration in a structured workflow. A within-subjects user study comparing SQLsynth with manual annotation and ChatGPT shows that SQLsynth significantly accelerates text-to-SQL data annotation, reduces cognitive load, and produces datasets that are more accurate, natural, and diverse. Our code is available at https://github.com/adobe/nl_sql_analyzer., Comment: Accepted by IUI'25

The quantum relative entropy is a fundamental quantity in quantum information science, characterizing the distinguishability between two quantum states. However, this quantity is not additive in general for correlated quantum states, necessitating regularization for precise characterization of the operational tasks of interest. Recently, we proposed the study of the regularized relative entropy between two sequences of sets of quantum states in [arXiv:2411.04035], which captures a general framework for a wide range of quantum information tasks. Here, we show that given suitable structural assumptions and efficient descriptions of the sets, the regularized relative entropy can be efficiently approximated within an additive error by a quantum relative entropy program of polynomial size. This applies in particular to the regularized relative entropy in adversarial quantum channel discrimination. Moreover, we apply the idea of efficient approximation to quantum resource theories. In particular, when the set of interest does not directly satisfy the required structural assumptions, it can be relaxed to one that does. This provides improved and efficient bounds for the entanglement cost of quantum states and channels, entanglement distillation and magic state distillation. Numerical results demonstrate improvements even for the first level of approximation., Comment: Expands on the computational aspects of arXiv:2411.04035 and contains the section "Application 4: efficient bounds for quantum resource theory" from arXiv:2411.04035v1

Large language models (LLMs) have achieved remarkable successes on various natural language tasks. However, recent studies have found that there are still significant challenges to the logical reasoning abilities of LLMs. This paper summarizes and categorizes the main challenges into two aspects: (1) Logical question answering, LLMs often fail to generate the correct answer within complex logical problem which requires sophisticated deductive, inductive or abductive reasoning given a collection of premises and constrains. (2) Logical consistency, LLMs are prone to producing responses contradicting themselves across different questions. For example, a state-of-the-art Macaw question-answering LLM answers Yes to both questions Is a magpie a bird? and Does a bird have wings? but answers No to Does a magpie have wings?. To facilitate this research direction, we comprehensively investigate the most cutting-edge methods and propose detailed taxonomies of these methods. Specifically, to accurately answer complex logic questions, previous methods can be categorized based on reliance on external solvers, prompts, pretraining, and fine-tuning. To avoid logical contradictions, we discuss concepts and solutions of various logical consistencies, including implication, negation, transitivity, factuality consistency, and their composites. In addition, we review commonly used benchmark datasets and evaluation metrics, and discuss promising research directions, such as extensions to modal logic to account for uncertainty, and efficient algorithms satisfying multiple logical consistencies simultaneously.

Kagome materials, which are composed of hexagons tiled with a shared triangle, have inspired enormous interest due to their unique structures and rich physical properties, but exploring superconducting material systems with new kagome structures is still an important research direction. Here, we predict a new type of kagome superconductors, MPd5 (M is a group IIA metal element), and identify that they exhibit coexistence of superconductivity and non-trivial topological properties. We uncover their phonon-mediated superconductivity by the density functional theory for superconductors (SCDFT), predicting the superconducting transition temperatures (Tc) of 2.64, 2.03, and 1.50 K for CaPd5, SrPd5, and BaPd5, respectively, which can be effectively tuned by external pressure and electron doping. The present results also demonstrate that MPd5 have various topological properties such as, CaPd5 shows topological non-trivial intersection near the Fermi level (EF). Our results indicate that the MPd5 materials can be an emerging material platform with rich exotic physics in their kagome structures, and render themselves excellent candidates for superconducting and advanced functional materials that could be utilized in topological quantum computing and information technology., Comment: 11 pages, 5 figures

Chemotaxis plays a significant role in numerous physiological processes. The Keller-Segel equation serves as a mathematical model for simulating the phenomenon of cell population aggregation under chemotaxis, possessing physical properties such as mass conservation, positivity of density, and energy dissipation. High-order linear and decoupled schemes for the parabolic-parabolic Keller-Segel chemotaxis model are proposed in this paper, which satisfy the three physical properties mentioned earlier. Firstly, by applying a logarithmic transformation, the Keller-Segel model is reformulated into its equivalent form that maintains the positivity of cell density regardless of the discrete scheme. Based on this equivalent system, we then propose high-order linear and decoupled numerical schemes using the backward differentiation formula (BDF). Furthermore, through the incorporation of a recovery technique and an energy-law preservation correction (EPC), we ensure that these schemes maintain mass conservation and preserve the original energy-law. Finally, we conduct a rigorous optimal error analysis for the numerical schemes under certain assumptions regarding the regularity of solutions, and some numerical experiments are also presented to demonstrate their effectiveness.

The increasing context window size in large language models (LLMs) has improved their ability to handle complex, long-text tasks. However, as the conversation rounds continue, it is required to store a large amount of KV cache in GPU memory, which significantly affects the efficiency and even availability of the model serving systems. This paper analyzes dialogue data from real users and discovers that the LLM inference manifests a watershed layer, after which the distribution of round-level attention shows notable similarity. We propose Round Attention, a novel round-level attention mechanism that only recalls and computes the KV cache of the most relevant rounds. The experiments show that our method saves 55\% memory usage without compromising model performance.

Multi-robot navigation in complex environments relies on inter-robot communication and mutual observations for coordination and situational awareness. This paper studies the multi-robot navigation problem in unknown environments with line-of-sight (LoS) connectivity constraints. While previous works are limited to known environment models to derive the LoS constraints, this paper eliminates such requirements by directly formulating the LoS constraints between robots from their real-time point cloud measurements, leveraging point cloud visibility analysis techniques. We propose a novel LoS-distance metric to quantify both the urgency and sensitivity of losing LoS between robots considering potential robot movements. Moreover, to address the imbalanced urgency of losing LoS between two robots, we design a fusion function to capture the overall urgency while generating gradients that facilitate robots' collaborative movement to maintain LoS. The LoS constraints are encoded into a potential function that preserves the positivity of the Fiedler eigenvalue of the robots' network graph to ensure connectivity. Finally, we establish a LoS-constrained exploration framework that integrates the proposed connectivity controller. We showcase its applications in multi-robot exploration in complex unknown environments, where robots can always maintain the LoS connectivity through distributed sensing and communication, while collaboratively mapping the unknown environment. The implementations are open-sourced at https://github.com/bairuofei/LoS_constrained_navigation., Comment: 8 pages, 9 figures, accepted by IEEE ICRA 2025

Traditional recommender systems usually take the user-platform paradigm, where users are directly exposed under the control of the platform's recommendation algorithms. However, the defect of recommendation algorithms may put users in very vulnerable positions under this paradigm. First, many sophisticated models are often designed with commercial objectives in mind, focusing on the platform's benefits, which may hinder their ability to protect and capture users' true interests. Second, these models are typically optimized using data from all users, which may overlook individual user's preferences. Due to these shortcomings, users may experience several disadvantages under the traditional user-platform direct exposure paradigm, such as lack of control over the recommender system, potential manipulation by the platform, echo chamber effects, or lack of personalization for less active users due to the dominance of active users during collaborative learning. Therefore, there is an urgent need to develop a new paradigm to protect user interests and alleviate these issues. Recently, some researchers have introduced LLM agents to simulate user behaviors, these approaches primarily aim to optimize platform-side performance, leaving core issues in recommender systems unresolved. To address these limitations, we propose a new user-agent-platform paradigm, where agent serves as the protective shield between user and recommender system that enables indirect exposure. To this end, we first construct four recommendation datasets, denoted as $\dataset$, along with user instructions for each record., Comment: WWW2025@HCRS