Search

This study addresses the important yet underexplored question of whether the Common Core State Standards in Mathematics, which emphasize critical thinking and problem-solving, as well as the computer-based assessments aligned with the Common Core, have facilitated or hindered learning for students with disabilities. By analyzing administrative data from a large county in California, we track mathematics achievement trends before and after the implementation of the Common Core. Our findings show a significant widening of the achievement disparity in mathematics between students with and without disabilities, suggesting that the Common Core and computerized assessments disproportionately affected students with disabilities.

Precise tracking of the growth in galaxy size and the evolution of merger fractions with redshift is vital for understanding the formation history of submillimeter galaxies (SMGs). This study investigates these evolutions over a broad redshift range ($1 < z \lesssim 6$), using a sample of 222 SMGs with a median redshift of $z = 2.61^{+0.89}_{-0.82}$ identified by ALMA and JCMT, enhanced by the advanced imaging capabilities of the JWST/NIRCam and MIRI. We find significant evolution in effective radii ($R_e$) in rest-frame V-band ($R_e \propto (1 + z)^{-0.87 \pm 0.08}$) and near-infrared (NIR) band ($R_e \propto (1 + z)^{-0.88 \pm 0.11}$), with the NIR size evolution resembling that of massive star-forming galaxies at lower redshift. Visual inspections reveal a major merger fraction of $24.3 \pm 3.7\%$ and an interaction fraction of up to $48.4 \pm 11.1\%$. The major merger fraction exhibits an increase from 14.7$\pm9.1$\% at $z = 1$ to 26.6$\pm 8.4$\% at $z = 3$, after which it remains approximately constant across the redshift range $3 < z < 6$. In contrast, the interaction fraction remains relatively stable across the range $2 < z < 5$. Our results indicate that late-stage major mergers are not the primary formation mechanism for SMGs at $z<3$, while interactions appear to play a significant role across the broader redshift range of $1

In this study, we present a comprehensive analysis of an unidentified point-like ultra-high-energy (UHE) $\gamma$-ray source, designated as 1LHAASO J1740+0948u, situated in the vicinity of the middle-aged pulsar PSR J1740+1000. The detection significance reached 17.1$\sigma$ (9.4$\sigma$) above 25$\,$TeV (100$\,$TeV). The source energy spectrum extended up to 300$\,$TeV, which was well fitted by a log-parabola function with $N0 = (1.93\pm0.23) \times 10^{-16} \rm{TeV^{-1}\,cm^{-2}\,s^{-2}}$, $\alpha = 2.14\pm0.27$, and $\beta = 1.20\pm0.41$ at E0 = 30$\,$TeV. The associated pulsar, PSR J1740+1000, resides at a high galactic latitude and powers a bow-shock pulsar wind nebula (BSPWN) with an extended X-ray tail. The best-fit position of the gamma-ray source appeared to be shifted by $0.2^{\circ}$ with respect to the pulsar position. As the (i) currently identified pulsar halos do not demonstrate such offsets, and (ii) centroid of the gamma-ray emission is approximately located at the extension of the X-ray tail, we speculate that the UHE $\gamma$-ray emission may originate from re-accelerated electron/positron pairs that are advected away in the bow-shock tail.

Low-bit model quantization for image super-resolution (SR) is a longstanding task that is renowned for its surprising compression and acceleration ability. However, accuracy degradation is inevitable when compressing the full-precision (FP) model to ultra-low bit widths (2~4 bits). Experimentally, we observe that the degradation of quantization is mainly attributed to the quantization of activation instead of model weights. In numerical analysis, the condition number of weights could measure how much the output value can change for a small change in the input argument, inherently reflecting the quantization error. Therefore, we propose CondiQuant, a condition number based low-bit post-training quantization for image super-resolution. Specifically, we formulate the quantization error as the condition number of weight metrics. By decoupling the representation ability and the quantization sensitivity, we design an efficient proximal gradient descent algorithm to iteratively minimize the condition number and maintain the output still. With comprehensive experiments, we demonstrate that CondiQuant outperforms existing state-of-the-art post-training quantization methods in accuracy without computation overhead and gains the theoretically optimal compression ratio in model parameters. Our code and model are released at https://github.com/Kai-Liu001/CondiQuant., Comment: 10 pages, 5 figures. Code and models are released at https://github.com/Kai-Liu001/CondiQuant

Large language models (LLMs) are widely used for natural language understanding and text generation. An LLM model relies on a time-consuming step called LLM decoding to generate output tokens. Several prior works focus on improving the performance of LLM decoding using parallelism techniques, such as batching and speculative decoding. State-of-the-art LLM decoding has both compute-bound and memory-bound kernels. Some prior works statically identify and map these different kernels to a heterogeneous architecture consisting of both processing-in-memory (PIM) units and computation-centric accelerators. We observe that characteristics of LLM decoding kernels (e.g., whether or not a kernel is memory-bound) can change dynamically due to parameter changes to meet user and/or system demands, making (1) static kernel mapping to PIM units and computation-centric accelerators suboptimal, and (2) one-size-fits-all approach of designing PIM units inefficient due to a large degree of heterogeneity even in memory-bound kernels. In this paper, we aim to accelerate LLM decoding while considering the dynamically changing characteristics of the kernels involved. We propose PAPI (PArallel Decoding with PIM), a PIM-enabled heterogeneous architecture that exploits dynamic scheduling of compute-bound or memory-bound kernels to suitable hardware units. PAPI has two key mechanisms: (1) online kernel characterization to dynamically schedule kernels to the most suitable hardware units at runtime and (2) a PIM-enabled heterogeneous computing system that harmoniously orchestrates both computation-centric processing units and hybrid PIM units with different computing capabilities. Our experimental results on three broadly-used LLMs show that PAPI achieves 1.8$\times$ and 11.1$\times$ speedups over a state-of-the-art heterogeneous LLM accelerator and a state-of-the-art PIM-only LLM accelerator, respectively., Comment: To appear in ASPLOS 2025

Embodied multimodal large models (EMLMs) have gained significant attention in recent years due to their potential to bridge the gap between perception, cognition, and action in complex, real-world environments. This comprehensive review explores the development of such models, including Large Language Models (LLMs), Large Vision Models (LVMs), and other models, while also examining other emerging architectures. We discuss the evolution of EMLMs, with a focus on embodied perception, navigation, interaction, and simulation. Furthermore, the review provides a detailed analysis of the datasets used for training and evaluating these models, highlighting the importance of diverse, high-quality data for effective learning. The paper also identifies key challenges faced by EMLMs, including issues of scalability, generalization, and real-time decision-making. Finally, we outline future directions, emphasizing the integration of multimodal sensing, reasoning, and action to advance the development of increasingly autonomous systems. By providing an in-depth analysis of state-of-the-art methods and identifying critical gaps, this paper aims to inspire future advancements in EMLMs and their applications across diverse domains., Comment: 81 pages, submitted to a journal for review

Multivariate Time Series Forecasting (MTSF) has long been a key research focus. Traditionally, these studies assume a fixed number of variables, but in real-world applications, Cyber-Physical Systems often expand as new sensors are deployed, increasing variables in MTSF. In light of this, we introduce a novel task, Expanding-variate Time Series Forecasting (EVTSF). This task presents unique challenges, specifically (1) handling inconsistent data shapes caused by adding new variables, and (2) addressing imbalanced spatio-temporal learning, where expanding variables have limited observed data due to the necessity for timely operation. To address these challenges, we propose STEV, a flexible spatio-temporal forecasting framework. STEV includes a new Flat Scheme to tackle the inconsistent data shape issue, which extends the graph-based spatio-temporal modeling architecture into 1D space by flattening the 2D samples along the variable dimension, making the model variable-scale-agnostic while still preserving dynamic spatial correlations through a holistic graph. We introduce a novel Spatio-temporal Focal Learning strategy that incorporates a negative filter to resolve potential conflicts between contrastive learning and graph representation, and a focal contrastive loss as its core to guide the framework to focus on optimizing the expanding variables. We benchmark EVTSF performance using three real-world datasets and compare it against three potential solutions employing SOTA MTSF models tailored for EVSTF. Experimental results show that STEV significantly outperforms its competitors, particularly on expanding variables. Notably, STEV, with only 5% of observations from the expanding period, is on par with SOTA MTSF models trained with complete observations. Further exploration of various expanding strategies underscores the generalizability of STEV in real-world applications.

We report the timing and spectral analyses upon of the type II X-ray bursts from the Rapid Burster (MXB 1730--335) observed by Insight-HXMT and Swift/XRT. By stacking the long-duration bursts, we find for the first time that the hard X-rays are lagging than the soft X-rays by 3 seconds. However, such a lag is not visible for the short-duration bursts, probably because of the poor statistics. For all bursts the energy spectrum is found to be non-thermal, thanks to the broad band coverage of Insight-HXMT. These findings put new insights into the type-II bursts and require a temporally showing-up corona for possible interpretation.

Object extraction and segmentation from remote sensing (RS) images is a critical yet challenging task in urban environment monitoring. Urban morphology is inherently complex, with irregular objects of diverse shapes and varying scales. These challenges are amplified by heterogeneity and scale disparities across RS data sources, including sensors, platforms, and modalities, making accurate object segmentation particularly demanding. While the Segment Anything Model (SAM) has shown significant potential in segmenting complex scenes, its performance in handling form-varying objects remains limited due to manual-interactive prompting. To this end, we propose UrbanSAM, a customized version of SAM specifically designed to analyze complex urban environments while tackling scaling effects from remotely sensed observations. Inspired by multi-resolution analysis (MRA) theory, UrbanSAM incorporates a novel learnable prompter equipped with a Uscaling-Adapter that adheres to the invariance criterion, enabling the model to capture multiscale contextual information of objects and adapt to arbitrary scale variations with theoretical guarantees. Furthermore, features from the Uscaling-Adapter and the trunk encoder are aligned through a masked cross-attention operation, allowing the trunk encoder to inherit the adapter's multiscale aggregation capability. This synergy enhances the segmentation performance, resulting in more powerful and accurate outputs, supported by the learned adapter. Extensive experimental results demonstrate the flexibility and superior segmentation performance of the proposed UrbanSAM on a global-scale dataset, encompassing scale-varying urban objects such as buildings, roads, and water.

Recent advances in Large Language Models (LLMs) have upgraded them from sophisticated text generators to autonomous agents capable of corporation and tool use in multi-agent systems (MASs). However, the robustness of these LLM-based MASs, especially under knowledge conflicts, remains unclear. In this paper, we design four comprehensive metrics to investigate the robustness of MASs when facing mild or task-critical knowledge conflicts. We first analyze mild knowledge conflicts introduced by heterogeneous agents and find that they do not harm system robustness but instead improve collaborative decision-making. Next, we investigate task-critical knowledge conflicts by synthesizing knowledge conflicts and embedding them into one of the agents. Our results show that these conflicts have surprisingly little to no impact on MAS robustness. Furthermore, we observe that MASs demonstrate certain self-repairing capabilities by reducing their reliance on knowledge conflicts and adopting alternative solution paths to maintain stability. Finally, we conduct ablation studies on the knowledge conflict number, agent number, and interaction rounds, finding that the self-repairing capability of MASs has intrinsic limits, and all findings hold consistently across various factors. Our code is publicly available at https://github.com/wbw625/MultiAgentRobustness., Comment: Working in progress

We present a model for analytically calculating gravitational lensing by self-interacting dark matter (SIDM) halos. Leveraging the universal behavior of SIDM halos during gravothermal evolution, we calibrate the lensing potential using a fluid simulation, normalizing the evolution time to align with established scenarios. From this potential, we derive explicit equations for the deflection angle and surface density profile, quantifying their deviations from numerical results. Our model builds on the parametric approach of arXiv:2305.16176, providing refinements in the deep core-collapse regime and enabling more comprehensive lensing studies. We explore characteristic lensing features, including critical curves and caustics, for SIDM halos in isolation and within a main halo, tracking their evolution through the gravothermal phase. We also examine signatures in the self-similar regime of core collapsed halos and highlight the role of baryonic effects in realistic halos. The efficiency of our model enables large-scale lensing studies, and we make our implementation publicly available on GitHub to support further research., Comment: 26 pages, 14 figures

Existing Large Vision-Language Models (LVLMs) can process inputs with context lengths up to 128k visual and text tokens, yet they struggle to generate coherent outputs beyond 1,000 words. We find that the primary limitation is the absence of long output examples during supervised fine-tuning (SFT). To tackle this issue, we introduce LongWriter-V-22k, a SFT dataset comprising 22,158 examples, each with multiple input images, an instruction, and corresponding outputs ranging from 0 to 10,000 words. Moreover, to achieve long outputs that maintain high-fidelity to the input images, we employ Direct Preference Optimization (DPO) to the SFT model. Given the high cost of collecting human feedback for lengthy outputs (e.g., 3,000 words), we propose IterDPO, which breaks long outputs into segments and uses iterative corrections to form preference pairs with the original outputs. Additionally, we develop MMLongBench-Write, a benchmark featuring six tasks to evaluate the long-generation capabilities of VLMs. Our 7B parameter model, trained with LongWriter-V-22k and IterDPO, achieves impressive performance on this benchmark, outperforming larger proprietary models like GPT-4o. Code and data: https://github.com/THU-KEG/LongWriter-V

Triton, a high-level Python-like language designed for building efficient GPU kernels, is widely adopted in deep learning frameworks due to its portability, flexibility, and accessibility. However, programming and parallel optimization still require considerable trial and error from Triton developers. Despite advances in large language models (LLMs) for conventional code generation, these models struggle to generate accurate, performance-optimized Triton code, as they lack awareness of its specifications and the complexities of GPU programming. More critically, there is an urgent need for systematic evaluations tailored to Triton. In this work, we introduce TritonBench, the first comprehensive benchmark for Triton operator generation. TritonBench features two evaluation channels: a curated set of 184 real-world operators from GitHub and a collection of operators aligned with PyTorch interfaces. Unlike conventional code benchmarks prioritizing functional correctness, TritonBench also profiles efficiency performance on widely deployed GPUs aligned with industry applications. Our study reveals that current state-of-the-art code LLMs struggle to generate efficient Triton operators, highlighting a significant gap in high-performance code generation. TritonBench will be available at https://github.com/thunlp/TritonBench.

Large language models (LLMs) have demonstrated remarkable proficiency in mainstream academic disciplines such as mathematics, physics, and computer science. However, human knowledge encompasses over 200 specialized disciplines, far exceeding the scope of existing benchmarks. The capabilities of LLMs in many of these specialized fields-particularly in light industry, agriculture, and service-oriented disciplines-remain inadequately evaluated. To address this gap, we present SuperGPQA, a comprehensive benchmark that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines. Our benchmark employs a novel Human-LLM collaborative filtering mechanism to eliminate trivial or ambiguous questions through iterative refinement based on both LLM responses and expert feedback. Our experimental results reveal significant room for improvement in the performance of current state-of-the-art LLMs across diverse knowledge domains (e.g., the reasoning-focused model DeepSeek-R1 achieved the highest accuracy of 61.82% on SuperGPQA), highlighting the considerable gap between current model capabilities and artificial general intelligence. Additionally, we present comprehensive insights from our management of a large-scale annotation process, involving over 80 expert annotators and an interactive Human-LLM collaborative system, offering valuable methodological guidance for future research initiatives of comparable scope.

Cold and cool gas (T $\leq 10^4$ K) in the circumgalactic medium (CGM) and its interaction with galaxies remain poorly understood. Simulations predict that cold gas flows into galaxies through cosmic filaments, determining the disk formation and galaxy evolution. The cold gas accretion modes in the CGM and their dependence on dark matter halo mass and redshift remain puzzling. Resolving the kiloparsec-scale kinematics and dynamics of cold gas interacting with the disk, dust, and metals in different environments is particularly lacking at z > 2. Here we report two disturbed cold gas structures traced by ultra-strong MgII absorbers (rest-frame equivalent width Wr > 2 \AA) exhibiting high kinematic velocities (> 500 km/s) and their environments at z ~ 4.9 and z ~ 2.6. Observations were conducted with VLT/MUSE, JWST/NIRCam, and ALMA to detect Lya and nebular emission lines, as well as dust continuum emission in the vicinity of these two absorbing gas structures. We identify two Lya emitters associated with a strong MgII absorber pair separated by ~1000 km/s at z~ 4.87. The pair exhibits relative differences in metallicity, dust content, and ionization states, suggesting internal metal and dust exchange within the ultra-large cold gas structure. For the strong MgII absorber at z = 2.5652$, we detect a dusty star-forming galaxy at a projected distance of $D = 38$ kpc. This galaxy exhibits prominent HeI, [SIII], and Paschen$\gamma$ lines, along with significant dust continuum. It has a star formation rate of ~ 121 +/- 33 $M_{\odot}$/yr and likely harbors a rotating disk. These findings tentatively suggest that cold gas at high redshifts plays a critical role in driving disk formation and actively participates in the transfer of metals and dust within the overdense regions of the CGM., Comment: 21 pages, 7 figures, 1 table in the main text, submitted

During the COVID-19 pandemic, built environments in dense urban settings become major sources of infection. This study tests the difference of demographics and surrounding built environments across high-, medium- and low-infection neighborhoods, to inform the high-risk areas in the city. We found that high-infection neighborhoods own a higher ratio of aged population than other neighborhoods on average. However, it shows no statistical difference in terms of population density. Additionally, high-infection neighborhoods are closer to high-risk built environments than the others. In a walking distance, they also can access more of the high-risk built environments except for the wholesale markets and shopping malls. These findings advise policy-makers to deploy social distancing measures in precision, regulating the access of high-risk facilities to mitigate the impacts of COVID-19.

Recently, LLMs have garnered increasing attention across academic disciplines for their potential as human digital twins, virtual proxies designed to replicate individuals and autonomously perform tasks such as decision-making, problem-solving, and reasoning on their behalf. However, current evaluations of LLMs primarily emphasize dialogue simulation while overlooking human behavior simulation, which is crucial for digital twins. To address this gap, we introduce BehaviorChain, the first benchmark for evaluating LLMs' ability to simulate continuous human behavior. BehaviorChain comprises diverse, high-quality, persona-based behavior chains, totaling 15,846 distinct behaviors across 1,001 unique personas, each with detailed history and profile metadata. For evaluation, we integrate persona metadata into LLMs and employ them to iteratively infer contextually appropriate behaviors within dynamic scenarios provided by BehaviorChain. Comprehensive evaluation results demonstrated that even state-of-the-art models struggle with accurately simulating continuous human behavior.

Quantum fluctuations are pivotal in driving quantum phase transitions, exemplified by the quantum melting of Wigner crystals into Fermi liquids in electron systems. However, their impact on superconducting systems near zero temperature, particularly in the superconductor-insulator/metal transition, remains poorly understood. In this study, through electric transport measurements on the two-dimensional (2D) superconductor (SnS)1.17NbS2, we demonstrate that quantum fluctuations induce vortex displacement from their mean position, leading to the quantum melting of vortex solid near zero temperature. Quantitative analysis reveals the magnetic field-induced anomalous metal originates from this quantum melting transition, with energy dissipation governed by quantum fluctuations-driven vortex displacements. Remarkably, further extending this analysis to various 2D superconductors yields the same results, and many properties of anomalous metal can be qualitatively understood within the framework of quantum melting. The connection between the quantum melting of vortex solids and dissipative anomalous metal opens a novel pathway towards understanding quantum phase transitions through vortex dynamics, providing new insights on both fields.

In unconventional superconductors such as cuprates and iron pnictides and chalcogenides, phase stiffness - a measure of the energy cost associated with superconducting phase variations - is on the same order of magnitude as the strength of Cooper pairing, translating to superconductivity governed by phase fluctuations. However, due to a lack of a direct experimental probe, there remains a fundamental gap in establishing microscopic picture between unconventional superconductivity and phase fluctuations. Here we show a vector current technique that allows for in-situ angle-resolved transport measurements, providing exclusive evidence suggesting an anisotropic nature of phase stiffness in infinite-layer nickelate superconductors. Pronounced anisotropy of in-plane resistance manifests itself in both normal and superconducting transition states, indicating crystal symmetry breaking. Remarkably, the electric conductivity of Nd0.8Sr0.2NiO2 peaks at 125{\deg} between the direction of the current and crystal principal axis, but this angle evolves to 160{\deg} near zero-resistance temperature. Further measurements reveal that the superconductivity is favored along a direction with minimized phase fluctuations, an orientation strikingly deviating from the symmetric direction imposed by both electronic anisotropy and the underlying crystal lattice. Identical measurements conducted on a prototypical cuprate superconductor yield consistent results, suggesting that this previously unknown behavior could be ubiquitous. By shielding insight into the contrasting anisotropy between electron fluid and superfluid, our findings provide clues for a unified framework for understanding unconventional superconductors

Comprehensible neural network explanations are foundations for a better understanding of decisions, especially when the input data are infused with malicious perturbations. Existing solutions generally mitigate the impact of perturbations through adversarial training, yet they fail to generate comprehensible explanations under unknown perturbations. To address this challenge, we propose AGAIN, a fActor GrAph-based Interpretable neural Network, which is capable of generating comprehensible explanations under unknown perturbations. Instead of retraining like previous solutions, the proposed AGAIN directly integrates logical rules by which logical errors in explanations are identified and rectified during inference. Specifically, we construct the factor graph to express logical rules between explanations and categories. By treating logical rules as exogenous knowledge, AGAIN can identify incomprehensible explanations that violate real-world logic. Furthermore, we propose an interactive intervention switch strategy rectifying explanations based on the logical guidance from the factor graph without learning perturbations, which overcomes the inherent limitation of adversarial training-based methods in defending only against known perturbations. Additionally, we theoretically demonstrate the effectiveness of employing factor graph by proving that the comprehensibility of explanations is strongly correlated with factor graph. Extensive experiments are conducted on three datasets and experimental results illustrate the superior performance of AGAIN compared to state-of-the-art baselines., Comment: The Thirteenth International Conference on Learning Representations

3D Semantic Scene Completion (SSC) provides comprehensive scene geometry and semantics for autonomous driving perception, which is crucial for enabling accurate and reliable decision-making. However, existing SSC methods are limited to capturing sparse information from the current frame or naively stacking multi-frame temporal features, thereby failing to acquire effective scene context. These approaches ignore critical motion dynamics and struggle to achieve temporal consistency. To address the above challenges, we propose a novel temporal SSC method FlowScene: Learning Temporal 3D Semantic Scene Completion via Optical Flow Guidance. By leveraging optical flow, FlowScene can integrate motion, different viewpoints, occlusions, and other contextual cues, thereby significantly improving the accuracy of 3D scene completion. Specifically, our framework introduces two key components: (1) a Flow-Guided Temporal Aggregation module that aligns and aggregates temporal features using optical flow, capturing motion-aware context and deformable structures; and (2) an Occlusion-Guided Voxel Refinement module that injects occlusion masks and temporally aggregated features into 3D voxel space, adaptively refining voxel representations for explicit geometric modeling. Experimental results demonstrate that FlowScene achieves state-of-the-art performance on the SemanticKITTI and SSCBench-KITTI-360 benchmarks.

We study the spectral and temporal properties of the black hole X-ray transient binary MAXI J1820+070 during the 2018 outburst with Insight-HXMT observations. The outburst of MAXI J1820+070 can be divided into three intervals. For the two intervals of the outburst, we find that low-energy (below 140 keV) photos lag high-energy (140-170 keV) ones, while in the decay of the outburst, high-energy photons lag low-energy photons, both with a time scale of the order of days. Based on these results, the canonical hysteresis effect of the 'q' shape in the hardness-intensity diagram can be reformed into a roughly linear shape by taking into account the lag corrections between different energy bands. Time analysis shows that the high-frequency break of hard X-rays, derived from the power density spectrum of the first interval of the outburst is, in general, larger and more variable than that of soft X-rays. The spectral fitting shows that the coverage fraction of the hard X-rays drops sharply at the beginning of the outburst to around 0.5, then increases slightly. The coverage fraction drops to roughly zero once the source steps into a soft state and increases gradually to unity when the source returns to a low hard state. We discuss the possible overall evolution scenario of corona hinted from these discoveries., Comment: 9 pages,10 figures. Published in MNRAS

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across a range of multimodal tasks. However, their inference efficiency is constrained by the large number of visual tokens processed during decoding. To address this challenge, we propose Per-Layer Per-Head Vision Token Pruning (PLPHP), a two-level fine-grained pruning method including Layer-Level Retention Rate Allocation and Head-Level Vision Token Pruning. Motivated by the Vision Token Re-attention phenomenon across decoder layers, we dynamically adjust token retention rates layer by layer. Layers that exhibit stronger attention to visual information preserve more vision tokens, while layers with lower vision attention are aggressively pruned. Furthermore, PLPHP applies pruning at the attention head level, enabling different heads within the same layer to independently retain critical context. Experiments on multiple benchmarks demonstrate that PLPHP delivers an 18% faster decoding speed and reduces the Key-Value Cache (KV Cache) size by over 50%, all at the cost of 0.46% average performance drop, while also achieving notable performance improvements in multi-image tasks. These results highlight the effectiveness of fine-grained token pruning and contribute to advancing the efficiency and scalability of LVLMs. Our source code will be made publicly available., Comment: 12 pages, 8 figures

In this work, we use the three-dimensional resistive MHD code NIMROD to investigate the impact of resistivity inhomogeneity on the sawtooth process of an reversed field pinch (RFP) plasma. The simulation employs a non-uniform resistivity profile similar to experiments, which monotonically increases from the core to the edge as the temperature decreases. The resistivity inhomogeneity introduces an additional electric field in the plasma, which accelerates the inward diffusion of magnetic flux and changing the self sustained reversal state, hence significantly enhances the dynamo effect and the sawtooth process in the RFP plasma.

Multi-turn instruction following capability constitutes a core competency of large language models (LLMs) in real-world applications. Existing evaluation benchmarks predominantly focus on fine-grained constraint satisfaction and domain-specific capability assessment, yet overlook the crucial structural dependency between dialogue turns that distinguishes multi-turn from single-turn interactions. This structural dependency not only reflects user intent but also establishes a second dimension for instruction following evaluation beyond constraint satisfaction. To address this gap, we propose StructFlowBench, a multi-turn instruction following benchmark with structural flow modeling. The benchmark innovatively defines a structural flow framework comprising six fundamental inter-turn relationships, which not only introduces novel structural constraints for model evaluation but also serves as generation parameters for creating customized dialogue flows tailored to specific scenarios. Adopting established LLM-based automatic evaluation methodologies, we conduct systematic evaluations of 13 leading open-source and closed-source LLMs. Experimental results reveal significant deficiencies in current models' comprehension of multi-turn dialogue structures. The code is available at \url{https://github.com/MLGroupJLU/StructFlowBench}., Comment: 18 pages, 8 figures, 8 tables

Optical modulators are essential building blocks for high-capacity optical communication and massively parallel computing. Among all types of optical modulators, travelling-wave Mach-Zehnder modulators (TW-MZMs) featuring high speed and efficiency are widely used, and have been developed on a variety of integrated material platforms. Existing methods to design and simulate TW-MZMs so far strongly rely on the peculiar material properties, and thus inevitably involve complicated electrical-circuit models. As a result, these methods diverge significantly. In addition, they become increasingly inefficient and inaccurate for TW-MZMs with extending length and levitating modulation speed, posing formidable challenges for millimeter-wave and terahertz operation. Here, we present an innovative perspective to understand and analyze high-speed TW-MZMs. Our perspective leverages nonlinear optics and complex band structures of RF photonic crystals, and is thus entirely electromagnetic-wave-based. Under this perspective, we showcase the design, optoelectronic simulation and experimental validation of high-speed TW-MZMs based on Si and LiNbO$_3$, and further demonstrate unambiguous advantages in simplicity, accuracy and efficiency over conventional methods. Our approach can essentially be applied to nearly any integrated material platform, including those based on semiconductors and electro-absorption materials. With high-frequency electrode designs and optoelectronic co-simulation, our approach facilitates the synergy and convergence of electronics and photonics, and offers a viable route to constructing future high-speed millimeter-wave and terahertz photonics and quantum systems.

Increasing test-time compute for LLMs shows promise across domains but remains underexplored in code generation, despite extensive study in math. In this paper, we propose S*, the first hybrid test-time scaling framework that substantially improves the coverage and selection accuracy of generated code. S* extends the existing parallel scaling paradigm with sequential scaling to push performance boundaries. It further leverages a novel selection mechanism that adaptively generates distinguishing inputs for pairwise comparison, combined with execution-grounded information to robustly identify correct solutions. We evaluate across 12 Large Language Models and Large Reasoning Model and show: (1) S* consistently improves performance across model families and sizes, enabling a 3B model to outperform GPT-4o-mini; (2) S* enables non-reasoning models to surpass reasoning models - GPT-4o-mini with S* outperforms o1-preview by 3.7% on LiveCodeBench; (3) S* further boosts state-of-the-art reasoning models - DeepSeek-R1-Distill-Qwen-32B with S* achieves 85.7% on LiveCodeBench, approaching o1 (high) at 88.5%. Code will be available under https://github.com/NovaSky-AI/SkyThought.

The integration of workflows with large language models (LLMs) enables LLM-based agents to execute predefined procedures, enhancing automation in real-world applications. Traditional rule-based methods tend to limit the inherent flexibility of LLMs, as their predefined execution paths restrict the models' action space, particularly when the unexpected, out-of-workflow (OOW) queries are encountered. Conversely, prompt-based methods allow LLMs to fully control the flow, which can lead to diminished enforcement of procedural compliance. To address these challenges, we introduce FlowAgent, a novel agent framework designed to maintain both compliance and flexibility. We propose the Procedure Description Language (PDL), which combines the adaptability of natural language with the precision of code to formulate workflows. Building on PDL, we develop a comprehensive framework that empowers LLMs to manage OOW queries effectively, while keeping the execution path under the supervision of a set of controllers. Additionally, we present a new evaluation methodology to rigorously assess an LLM agent's ability to handle OOW scenarios, going beyond routine flow compliance tested in existing benchmarks. Experiments on three datasets demonstrate that FlowAgent not only adheres to workflows but also effectively manages OOW queries, highlighting its dual strengths in compliance and flexibility. The code is available at https://github.com/Lightblues/FlowAgent., Comment: 8 pages

Direct Preference Optimization (DPO) has gained attention as an efficient alternative to reinforcement learning from human feedback (RLHF) for aligning large language models (LLMs) with human preferences. Despite its advantages, DPO suffers from a length bias, generating responses longer than those from the reference model. Existing solutions like SimPO and SamPO address this issue but uniformly treat the contribution of rewards across sequences, overlooking temporal dynamics. To this end, we propose an enhanced preference optimization method that incorporates a temporal decay factor controlled by a gamma parameter. This dynamic weighting mechanism adjusts the influence of each reward based on its position in the sequence, prioritizing earlier tokens that are more critical for alignment. By adaptively focusing on more relevant feedback, our approach mitigates overfitting to less pertinent data and remains responsive to evolving human preferences. Experimental results on several benchmarks show that our approach consistently outperforms vanilla DPO by 5.9-8.8 points on AlpacaEval 2 and 3.3-9.7 points on Arena-Hard across different model architectures and sizes. Furthermore, additional experiments on mathematical and reasoning benchmarks (MMLU, GSM8K, and MATH) confirm that our method enhances performance without compromising general capabilities. Our codebase would be available at \url{https://github.com/LotuSrc/D2PO}., Comment: Accepted by ICLR 2025

Generative Foundation Models (GenFMs) have emerged as transformative tools. However, their widespread adoption raises critical concerns regarding trustworthiness across dimensions. This paper presents a comprehensive framework to address these challenges through three key contributions. First, we systematically review global AI governance laws and policies from governments and regulatory bodies, as well as industry practices and standards. Based on this analysis, we propose a set of guiding principles for GenFMs, developed through extensive multidisciplinary collaboration that integrates technical, ethical, legal, and societal perspectives. Second, we introduce TrustGen, the first dynamic benchmarking platform designed to evaluate trustworthiness across multiple dimensions and model types, including text-to-image, large language, and vision-language models. TrustGen leverages modular components--metadata curation, test case generation, and contextual variation--to enable adaptive and iterative assessments, overcoming the limitations of static evaluation methods. Using TrustGen, we reveal significant progress in trustworthiness while identifying persistent challenges. Finally, we provide an in-depth discussion of the challenges and future directions for trustworthy GenFMs, which reveals the complex, evolving nature of trustworthiness, highlighting the nuanced trade-offs between utility and trustworthiness, and consideration for various downstream applications, identifying persistent challenges and providing a strategic roadmap for future research. This work establishes a holistic framework for advancing trustworthiness in GenAI, paving the way for safer and more responsible integration of GenFMs into critical applications. To facilitate advancement in the community, we release the toolkit for dynamic evaluation.

Aligning small language models (SLMs) with human values typically involves distilling preference knowledge from large language models (LLMs). However, existing distillation methods model preference knowledge in teacher LLMs by comparing pairwise responses, overlooking the extent of difference between responses. This limitation hinders student SLMs from capturing the nuanced preferences for multiple responses. In this paper, we propose a Preference-Aligned Distillation (PAD) framework, which models teacher's preference knowledge as a probability distribution over all potential preferences, thereby providing more nuanced supervisory signals. Our insight in developing PAD is rooted in the demonstration that language models can serve as reward functions, reflecting their intrinsic preferences. Based on this, PAD comprises three key steps: (1) sampling diverse responses using high-temperature; (2) computing rewards for both teacher and student to construct their intrinsic preference; and (3) training the student's intrinsic preference distribution to align with the teacher's. Experiments on four mainstream alignment benchmarks demonstrate that PAD consistently and significantly outperforms existing approaches, achieving over 20\% improvement on AlpacaEval 2 and Arena-Hard, indicating superior alignment with human preferences. Notably, on MT-Bench, using the \textsc{Gemma} model family, the student trained by PAD surpasses its teacher, further validating the effectiveness of our PAD., Comment: Under review

End-to-end autonomous driving, which directly maps raw sensor inputs to low-level vehicle controls, is an important part of Embodied AI. Despite successes in applying Multimodal Large Language Models (MLLMs) for high-level traffic scene semantic understanding, it remains challenging to effectively translate these conceptual semantics understandings into low-level motion control commands and achieve generalization and consensus in cross-scene driving. We introduce Sce2DriveX, a human-like driving chain-of-thought (CoT) reasoning MLLM framework. Sce2DriveX utilizes multimodal joint learning from local scene videos and global BEV maps to deeply understand long-range spatiotemporal relationships and road topology, enhancing its comprehensive perception and reasoning capabilities in 3D dynamic/static scenes and achieving driving generalization across scenes. Building on this, it reconstructs the implicit cognitive chain inherent in human driving, covering scene understanding, meta-action reasoning, behavior interpretation analysis, motion planning and control, thereby further bridging the gap between autonomous driving and human thought processes. To elevate model performance, we have developed the first extensive Visual Question Answering (VQA) driving instruction dataset tailored for 3D spatial understanding and long-axis task reasoning. Extensive experiments demonstrate that Sce2DriveX achieves state-of-the-art performance from scene understanding to end-to-end driving, as well as robust generalization on the CARLA Bench2Drive benchmark.

This paper studies the thermo-poroelasticity model. By introducing an intermediate variable, we transform the original three-field model into a four-field model. Building upon this four-field model, we present both a coupled finite element method and a decoupled iterative finite element method. We prove the stability and optimal convergence of the coupled finite element method. Furthermore, we establish the convergence of the decoupled iterative method. This paper focuses primarily on analyzing the iterative decoupled algorithm. It demonstrates that the algorithm's convergence does not require any additional assumptions about physical parameters or stabilization parameters. Numerical results are provided to demonstrate the effectiveness and theoretical validity of these new methods., Comment: submitted to a journal

Current observations indicate that stars with higher rotation rates appear to maintain more surface lithium, and the majority of lithium-rich giants are indeed red clump stars. Hence, we investigate the mechanisms behind lithium enrichment in rotating red clump stars and the pathways to forming lithium-rich red clump stars. Meridional circulation is prevalent in the radiative zone of rotating giants. We model its radial mixing as a diffusion process and derive the corresponding diffusion coefficient based on its material transfer effect. Due to uncertainties in numerical calculations, we consider an average diffusion effect. Additionally, certain limiting conditions for the radial velocity of meridional circulation are incorporated. With varying input rotation velocities, we simulate the lithium evolution for red clump stars with this model. Our results indicate that the material transfer effect due to meridional circulation can efficiently transport beryllium, produced by H burning, into the convective envelope. This meridional circulation can lead to lithium enrichment, with a maximum lithium abundance increment approaching \(3.0\,\rm dex\). Consequently, it is capable of forming both lithium-rich and super lithium-rich red clump stars. The degree of lithium enrichment exhibits a strong positive correlation with the rotation velocity, i.e., faster red clump stars show more surface lithium. Furthermore, our models indicate that lithium-rich red clump stars are relatively young (\(\sim 10^6\,\rm yr\)), which aligns with observation evidences., Comment: Accepted by ApJ Letters

The evolution and sudden change of steady interaction structures is numerically studied with the fore wedge angle theta_1 in a low enthalpy hypersonic double wedge configuration. It particularly focuses on the conditions of Swantek and Austin's experiments where Ma=7, and h_0=2 MJ/kg but with a reduced Reynolds number (Re). The sudden structural change indicates that when theta_1 reaches a critical value, minor angular variations can trigger a discontinuous transformation in flow structures. The analysis is based on the laminar Navier-Stokes equations, using ideal gas and non-equilibrium gas models. Under the condition of Re=1E5/m, detailed numerical simulations are conducted as theta_1 varies over 0 deg-40 deg. This study yields the following findings: (a) The upper and lower boundaries of theta_1 for the onset of unsteady flow are identified. When theta_1 lies outside these boundaries, the flow remains steady. (b) As theta_1 increases, the interaction patterns evolve sequentially, progressing from Type VI through Type VI->V, Type III, Type IV_r, and ultimately to a flow dominated solely by a bow shock. This evolution defines the boundaries between different interaction patterns and provides a comprehensive understanding of their progression with theta_1. Sudden structural changes occur during the transitions from Type III to Type IV_r and from Type IV_r to a bow shock-dominated flow. In addition, a comparative study is performed through shock polar analysis to compare its predictions with computational results. (c) An unconventional reflection pattern of the transmitted shock over the separation zone, called Type III_r, is observed in non-equilibrium gas flows, which differs from classical interaction patterns. (d) The aerodynamic distribution of wall properties under various interactions is obtained, indicating distinct features before and after the sudden structural change.

Graph neural networks have been demonstrated as a powerful paradigm for effectively learning graph-structured data on the web and mining content from it.Current leading graph models require a large number of labeled samples for training, which unavoidably leads to overfitting in few-shot scenarios. Recent research has sought to alleviate this issue by simultaneously leveraging graph learning and meta-learning paradigms. However, these graph meta-learning models assume the availability of numerous meta-training tasks to learn transferable meta-knowledge. Such assumption may not be feasible in the real world due to the difficulty of constructing tasks and the substantial costs involved. Therefore, we propose a SiMple yet effectIve approach for graph few-shot Learning with fEwer tasks, named SMILE. We introduce a dual-level mixup strategy, encompassing both within-task and across-task mixup, to simultaneously enrich the available nodes and tasks in meta-learning. Moreover, we explicitly leverage the prior information provided by the node degrees in the graph to encode expressive node representations. Theoretically, we demonstrate that SMILE can enhance the model generalization ability. Empirically, SMILE consistently outperforms other competitive models by a large margin across all evaluated datasets with in-domain and cross-domain settings. Our anonymous code can be found here., Comment: WWW25

Large multimodal models (LMMs) have shown impressive capabilities in a wide range of visual tasks. However, they often struggle with fine-grained visual reasoning, failing to identify domain-specific objectives and provide justifiable explanations for their predictions. To address this, we propose a novel visual rejection sampling framework to improve the cognition and explainability of LMMs using self-synthesized data. Specifically, visual fine-tuning requires images, queries, and target answers. Our approach begins by synthesizing interpretable answers that include human-verifiable visual features. These features are based on expert-defined concepts, carefully selected based on their alignment with the image content. After each round of fine-tuning, we apply a reward model-free filtering mechanism to select the highest-quality interpretable answers for the next round of tuning. This iterative process of data synthesis and fine-tuning progressively improves the model's ability to generate accurate and reasonable explanations. Experimental results demonstrate the effectiveness of our method in improving both the accuracy and explainability of specialized visual classification tasks., Comment: Accepted by ICLR 2025. Code: https://github.com/sycny/SelfSynthX

Current recommendation systems powered by large language models (LLMs) often underutilize their reasoning capabilities due to a lack of explicit logical structuring. To address this limitation, we introduce CoT-Rec, a framework that integrates Chain-of-Thought (CoT) reasoning into LLM-driven recommendations by incorporating two crucial processes: user preference analysis and item perception evaluation. CoT-Rec operates in two key phases: (1) personalized data extraction, where user preferences and item perceptions are identified, and (2) personalized data application, where this information is leveraged to refine recommendations. Our experimental analysis demonstrates that CoT-Rec improves recommendation accuracy by making better use of LLMs' reasoning potential. The implementation is publicly available at https://anonymous.4open.science/r/CoT-Rec., Comment: 7 pages, under review

Large Language Model (LLM)-based user agents have emerged as a powerful tool for improving recommender systems by simulating user interactions. However, existing methods struggle with cross-domain scenarios due to inefficient memory structures, leading to irrelevant information retention and failure to account for social influence factors such as popularity. To address these limitations, we introduce AgentCF++, a novel framework featuring a dual-layer memory architecture and a two-step fusion mechanism to filter domain-specific preferences effectively. Additionally, we propose interest groups with shared memory, allowing the model to capture the impact of popularity trends on users with similar interests. Through extensive experiments on multiple cross-domain datasets, AgentCF++ demonstrates superior performance over baseline models, highlighting its effectiveness in refining user behavior simulation for recommender systems. Our code is available at https://anonymous.4open.science/r/AgentCF-plus., Comment: 6 pages, under review

Recommender systems often suffer from popularity bias, where frequently interacted items are overrepresented in recommendations. This bias stems from propensity factors influencing training data, leading to imbalanced exposure. In this paper, we introduce a Fair Sampling (FS) approach to address this issue by ensuring that both users and items are selected with equal probability as positive and negative instances. Unlike traditional inverse propensity score (IPS) methods, FS does not require propensity estimation, eliminating errors associated with inaccurate calculations. Our theoretical analysis demonstrates that FS effectively neutralizes the influence of propensity factors, achieving unbiased learning. Experimental results validate that FS outperforms state-of-the-art methods in both point-wise and pair-wise recommendation tasks, enhancing recommendation fairness without sacrificing accuracy. The implementation is available at https://anonymous.4open.science/r/Fair-Sampling., Comment: 6 pages, under review

Despite significant advancements in traditional syntactic communications based on Shannon's theory, these methods struggle to meet the requirements of 6G immersive communications, especially under challenging transmission conditions. With the development of generative artificial intelligence (GenAI), progress has been made in reconstructing videos using high-level semantic information. In this paper, we propose a scalable generative video semantic communication framework that extracts and transmits semantic information to achieve high-quality video reconstruction. Specifically, at the transmitter, description and other condition signals (e.g., first frame, sketches, etc.) are extracted from the source video, functioning as text and structural semantics, respectively. At the receiver, the diffusion-based GenAI large models are utilized to fuse the semantics of the multiple modalities for reconstructing the video. Simulation results demonstrate that, at an ultra-low channel bandwidth ratio (CBR), our scheme effectively captures semantic information to reconstruct videos aligned with human perception under different signal-to-noise ratios. Notably, the proposed ``First Frame+Desc." scheme consistently achieves CLIP score exceeding 0.92 at CBR = 0.0057 for SNR > 0 dB. This demonstrates its robust performance even under low SNR conditions.

Large language models (LLMs) can prove mathematical theorems formally by generating proof steps (\textit{a.k.a.} tactics) within a proof system. However, the space of possible tactics is vast and complex, while the available training data for formal proofs is limited, posing a significant challenge to LLM-based tactic generation. To address this, we introduce a neuro-symbolic tactic generator that synergizes the mathematical intuition learned by LLMs with domain-specific insights encoded by symbolic methods. The key aspect of this integration is identifying which parts of mathematical reasoning are best suited to LLMs and which to symbolic methods. While the high-level idea of neuro-symbolic integration is broadly applicable to various mathematical problems, in this paper, we focus specifically on Olympiad inequalities (Figure~1). We analyze how humans solve these problems and distill the techniques into two types of tactics: (1) scaling, handled by symbolic methods, and (2) rewriting, handled by LLMs. In addition, we combine symbolic tools with LLMs to prune and rank the proof goals for efficient proof search. We evaluate our framework on 161 challenging inequalities from multiple mathematics competitions, achieving state-of-the-art performance and significantly outperforming existing LLM and symbolic approaches without requiring additional training data., Comment: Published as a conference paper at ICLR 2025. Code is available at https://github.com/Lizn-zn/NeqLIPS/

In this work, we study the masses of the $1S$-, $1P$-, $1D$-, $2S$-, and $2P$-wave $\Omega_c$ baryon states within a semi-relativistic constituent quark potential model. Furthermore, the strong decay properties for the excited $\Omega_c$ states are evaluated by using the masses and wave functions obtained from the potential model. In a unified framework, we provide a reasonable explanation of both the masses and widths for the newly observed $\Omega_c$ resonances $\Omega_c(3000)$, $\Omega_c(3050)$, $\Omega_c(3065)$, $\Omega_c(3090)$, $\Omega_c(3120)$, $\Omega_c(3185)$, and $\Omega_c(3327)$. Based on the obtained decay properties and mass spectrum, we further suggest optimal channels and mass regions to find the missing $\Omega_c$ resonances. We expect our study can provide a useful reference for establishing the $\Omega_c$ spectrum., Comment: 15 pages, 5 figures

Integrated sensing and communication (ISAC) plays a crucial role in the Internet of Vehicles (IoV), serving as a key factor in enhancing driving safety and traffic efficiency. To address the security challenges of the confidential information transmission caused by the inherent openness nature of wireless medium, different from current physical layer security (PLS) methods, which depends on the \emph{additional communication interference} costing extra power resources, in this paper, we investigate a novel PLS solution, under which the \emph{inherent radar sensing interference} of the vehicles is utilized to secure wireless communications. To measure the performance of PLS methods in ISAC-based IoV systems, we first define an improved security performance metric called by transmission reliability and sensing accuracy based secrecy rate (TRSA\_SR), and derive closed-form expressions of connection outage probability (COP), secrecy outage probability (SOP), success ranging probability (SRP) for evaluating transmission reliability, security and sensing accuracy, respectively. Furthermore, we formulate an optimization problem to maximize the TRSA\_SR by utilizing radar sensing interference and joint design of the communication duration, transmission power and straight trajectory of the legitimate transmitter. Finally, the non-convex feature of formulated problem is solved through the problem decomposition and alternating optimization. Simulations indicate that compared with traditional PLS methods obtaining a non-positive STC, the proposed method achieves a secrecy rate of 3.92bps/Hz for different settings of noise power.

Evaluating models on large benchmarks is very resource-intensive, especially during the period of rapid model evolution. Existing efficient evaluation methods estimate the performance of target models by testing them only on a small and static coreset of the benchmark, which is derived from the publicly available evaluation results of source models. These methods rely on the assumption that target models have high prediction consistency with source models. However, we demonstrate that it doesn't generalize well in practice. To alleviate the inconsistency issue, we present TailoredBench, a method that conducts customized evaluation tailored to each target model. Specifically, a Global-coreset is first constructed as a probe to identify the most consistent source models for each target model with an adaptive source model selection strategy. Afterwards, a scalable K-Medoids clustering algorithm is proposed to extend the Global-coreset to a tailored Native-coreset for each target model. According to the predictions on Native-coresets, we obtain the performance of target models on the whole benchmark with a calibrated estimation strategy. Comprehensive experiments on 5 benchmarks across over 300 models demonstrate that compared to best performing baselines, TailoredBench achieves an average reduction of 31.4% in MAE of accuracy estimates under the same inference budgets, showcasing strong effectiveness and generalizability.

Using $(2712\pm14)\times10^6$ $\psi(3686)$ events collected with the BESIII detector, we perform the first amplitude analysis of the radiative decay $\psi(3686)\to \gamma K_S^0 K_S^0$ within the mass region $M_{K_S^0 K_S^0 }<2.8$ GeV/$c^2$. Employing a one-channel K-matrix approach for the description of the dynamics of the $K^0_S K^0_S$ system, the data sample is well described with four poles for the $f_0$-wave and three poles for the $f_2$-wave. The determined pole positions are consistent with those of well-established resonance states. The observed $f_0$ and $f_{2}$ states are found to be qualitatively consistent with those produced in radiative $J/\psi$ decays, indicating the similarity between the two charmonium states in their radiative decays., Comment: 20 pages, 4 figures, submitted to JHEP

Despite the rapid progress of large language models (LLMs), their length-controllable text generation (LCTG) ability remains below expectations, posing a major limitation for practical applications. Existing methods mainly focus on end-to-end training to reinforce adherence to length constraints. However, the lack of decomposition and targeted enhancement of LCTG sub-abilities restricts further progress. To bridge this gap, we conduct a bottom-up decomposition of LCTG sub-abilities with human patterns as reference and perform a detailed error analysis. On this basis, we propose MarkerGen, a simple-yet-effective plug-and-play approach that:(1) mitigates LLM fundamental deficiencies via external tool integration;(2) conducts explicit length modeling with dynamically inserted markers;(3) employs a three-stage generation scheme to better align length constraints while maintaining content quality. Comprehensive experiments demonstrate that MarkerGen significantly improves LCTG across various settings, exhibiting outstanding effectiveness and generalizability.

Recent advances in large language models (LLMs) have showcased exceptional performance in long-context tasks, while facing significant inference efficiency challenges with limited GPU memory. Existing solutions first proposed the sliding-window approach to accumulate a set of historical \textbf{key-value} (KV) pairs for reuse, then further improvements selectively retain its subsets at each step. However, due to the sparse attention distribution across a long context, it is hard to identify and recall relevant KV pairs, as the attention is distracted by massive candidate pairs. Additionally, we found it promising to select representative tokens as probe-Query in each sliding window to effectively represent the entire context, which is an approach overlooked by existing methods. Thus, we propose \textbf{ActQKV}, a training-free, \textbf{Act}ivation-aware approach that dynamically determines probe-\textbf{Q}uery and leverages it to retrieve the relevant \textbf{KV} pairs for inference. Specifically, ActQKV monitors a token-level indicator, Activation Bias, within each context window, enabling the proper construction of probe-Query for retrieval at pre-filling stage. To accurately recall the relevant KV pairs and minimize the irrelevant ones, we design a dynamic KV cut-off mechanism guided by information density across layers at the decoding stage. Experiments on the Long-Bench and $\infty$ Benchmarks demonstrate its state-of-the-art performance with competitive inference quality and resource efficiency.

Previous studies have established that quasiperiodic lattice models with unbounded potentials can exhibit localized and multifractal states, yet preclude the existence of extended states. In this work, we introduce a quasiperiodic system that incorporates both unbounded potentials and unbounded hopping amplitudes, where extended states emerge as a direct consequence of the unbounded hopping terms overcoming the localization constraints imposed by the unbounded potential, thereby facilitating enhanced particle transport. By using Avila's global theory, we derive analytical expressions for the phase boundaries, with exact results aligning closely with numerical simulations.Intriguingly, we uncover a hidden self-duality in the proposed model by establishing a mapping to the Aubry-Andr\'e model, revealing a profound structural connection between these systems., Comment: 7 pages, 3 figures

Potentially Hazardous Asteroids (PHAs), a special subset of Near-Earth Objects, are both dangerous and scientifically valuable. PHAs that truly undergo close approaches with the Earth (dubbed CAPHAs) are of particular interest and extensively studied. The concept and study of CAPHA can be extended to other Solar system planets, which have significant implications for future planet-based observations and explorations. In this work, we conduct numerical simulations that incorporate the Yarkovsky effect to study the transformation of main belt asteroids into CAPHAs of terrestrial planets, using precise nominal timesteps, especially to ensure the reliability of the results for Mercury and Venus. Our simulations predict a total of 1893 Mercury-CAPHAs, 3014 Venus-CAPHAs, 3791 Earth-CAPHAs and 18066 Mars-CAPHAs, with an occurrence frequency of about 1, 9, 15 and 66 per year, respectively. The values for Mars-CAPHAs are consistent with our previous work, which were based on simulations with a larger nominal timestep. The predicted occurrence frequency and velocity distribution of Earth-CAPHAs are in reasonable agreement with the observed population of Earth-CAPHAs. We also find that certain asteroids can be caught in close approach with different planets at different times, raising an interesting possibility of using them as transportation between terrestrial planets in the future., Comment: 8 pages, 4 figures, accepted for publication in MNRAS. arXiv admin note: text overlap with arXiv:2405.02614