arXiv subjects

Jun Chen

Publications and source records attributed to Jun Chen.

At least 109 records · Page 6Linked to original sources

arXiv2026

Feedback Does Not Increase the Capacity of Approximately Memoryless Surjective POST Channels

We study a class of finite-state channels, known as POST channels, in which the previous channel output serves as the current state. A POST channel is deemed approximately memoryless when the state-dependent transition matrices are sufficiently close to one another. For this family of channels, under a surjectivity condition on the associated memoryless reference channel, we show that the feedback capacity coincides with the non-feedback capacity. Consequently, for almost all approximately memoryless POST channels whose input alphabet size is no smaller than the output alphabet size, feedback provides no capacity gain. This result extends Shannon's classical theorem on discrete memoryless channels and demonstrates that the phenomenon holds well beyond the strictly memoryless case.

cs.IT↗

arXiv2026

KramaBench: A Benchmark for AI Systems on Data-to-Insight Pipelines over Data Lakes

Discovering insights from a real-world data lake potentially containing unclean, semi-structured, and unstructured data requires a variety of data processing tasks, ranging from extraction and cleaning to integration, analysis, and modeling. This process often also demands domain knowledge and project-specific insight. While AI models have shown remarkable results in reasoning and code generation, their abilities to design and execute complex pipelines that solve these data-lake-to-insight challenges remain unclear. We introduce KramaBench which consists of 104 manually curated and solved challenges spanning 1700 files, 24 data sources, and 6 domains. KramaBench focuses on testing the end-to-end capabilities of AI systems to solve challenges which require automated orchestration of different data tasks. KramaBench also features a comprehensive evaluation framework assessing the pipeline design and individual data task implementation abilities of AI systems. We evaluate 8 LLMs using our single-agent reference framework DS-Guru, alongside both open- and closed-source single- and multi-agent systems, and find that while current agentic systems may handle isolated data-science tasks and generate plausible draft pipelines, they struggle with producing working end-to-end pipelines. On KramaBench, the best system reaches only 55% end-to-end accuracy in the full data-lake setting. Even with perfect retrieval, the accuracy tops out at 62%. Leading LLMs can identify up to 42% of important data tasks but can only fully implement 20% of individual data tasks. Our code, reference framework, and data are available at https://github.com/mitdbg/KramaBench.

cs.DB↗

arXiv2026

MultiHaystack: Benchmarking Multimodal Retrieval and Reasoning over 40K Images, Videos, and Documents

Multimodal large language models (MLLMs) achieve strong performance on benchmarks that evaluate text, image, or video understanding separately. However, these settings do not assess a critical real-world requirement, which involves retrieving relevant evidence from large, heterogeneous multimodal corpora prior to reasoning. Most existing benchmarks restrict retrieval to small, single-modality candidate sets, substantially simplifying the search space and overstating end-to-end reliability. To address this gap, we introduce MultiHaystack, the first benchmark designed to evaluate both retrieval and reasoning under large-scale, cross-modal conditions. MultiHaystack comprises over 46,000 multimodal retrieval candidates across documents, images, and videos, along with 747 open yet verifiable questions. Each question is grounded in a unique validated evidence item within the retrieval pool, requiring evidence localization across modalities and fine-grained reasoning. In our study, we find that models perform competitively when provided with the corresponding evidence, but their performance drops sharply when required to retrieve that evidence from the full corpus. Additionally, even the strongest retriever, E5-V, achieves only 40.8% Recall@1, while state-of-the-art MLLMs such as GPT-5 experience a significant drop in reasoning accuracy from 80.86% when provided with the corresponding evidence to 51.4% under top-5 retrieval. These results indicate that multimodal retrieval over heterogeneous pools remains a primary bottleneck for MLLMs, positioning MultiHaystack as a valuable testbed that highlights underlying limitations obscured by small-scale evaluations and promotes retrieval-centric advances in multimodal systems.

cs.CV↗

arXiv2026

W4A16 Mixed-Precision Matrix Multiplication on Decoupled Architecture: Kernel Design and Memory Bottleneck Analysis for Ascend NPUs

As Large Language Models (LLMs) scale, weight-only quantization (W4A16: 4-bit weights, 16-bit activations) becomes critical for reducing memory footprint with minimal accuracy loss. However, its efficient deployment on Huawei's Ascend 910 Neural Processing Unit (NPU) is challenging due to limited native mixed-precision support and the accelerator's decoupled compute architecture. To enable quantization on such architecture, we present the first practical W4A16 matrix multiplication kernel tailored for the Ascend 910 NPU. Our design leverages vector cores for on-the-fly INT4-to-FP16 dequantization, cube cores for high-throughput GEMM, and Split-K parallelization to mitigate memory latency. Performance evaluations across diverse matrix shapes and batch sizes show our method outperforms data-parallel approaches when K >> N, a typical scenario in LLM decoding. Specially, our method can achieve a speedup ranging from 1.01x to 1.74x. In addition, our profile reveals the primary bottleneck is not dequantization compution itself, but extra global memory transfer for the weight, making W4A16 only reaching a maximum speedup of 1.48x over native FP16xFP16 matrix multiplication in PyTorch. In the long run, our method lays a solid foundation and provides insightful views for the efficient deployment of quantized large language models on various domain-specific accelerators.

cs.DC↗

arXiv2026

CiteLLM: An Agentic Platform for Trustworthy Scientific Reference Discovery

Large language models (LLMs) have created new opportunities to enhance the efficiency of scholarly activities; however, challenges persist in the ethical deployment of AI assistance, including (1) the trustworthiness of AI-generated content, (2) preservation of academic integrity and intellectual property, and (3) protection of information privacy. In this work, we present CiteLLM, a specialized agentic platform designed to enable trustworthy reference discovery for grounding author-drafted claims and statements. The system introduces a novel interaction paradigm by embedding LLM utilities directly within the LaTeX editor environment, ensuring a seamless user experience and no data transmission outside the local system. To guarantee hallucination-free references, we employ dynamic discipline-aware routing to retrieve candidates exclusively from trusted web-based academic repositories, while leveraging LLMs solely for generating context-aware search queries, ranking candidates by relevance, and validating and explaining support through paragraph-level semantic matching and an integrated chatbot. Evaluation results demonstrate the superior performance of the proposed system in returning valid and highly usable references.

cs.CL↗

arXiv2026

On Linear Power Control Policies for Energy Harvesting Communications

This paper studies optimal linear power control for battery-limited energy harvesting communications. It provides a systematic analysis of linear power control policies, covering the greedy and fixed-fraction policies as special cases. Three optimality notions are introduced: the maximin optimal linear policy for a given battery capacity and mean-to-capacity ratio (MCR), and two capacity-agnostic policies that minimize the nominal additive gap and maximize the nominal multiplicative factor, respectively. Except the capacity-agnostic additive-gap optimal linear policy, which coincides with the fixed-fraction policy, the other two optimal linear policies are novel and constitute the main contributions of this paper. It is shown, among others, that the worst nominal multiplicative factor for both novel policies is approximately 0.6530, a substantial improvement over the fixed-fraction policy's value of 0.5. Simulations show that under quasi-static fading, the maximin optimal linear policy performs comparable to the maximin optimal policy (the top-performing policy), while the capacity-agnostic multiplicative-factor optimal linear policy performs slightly worse; nevertheless, both novel policies significantly outperform the fixed-fraction policy in the low-to-medium signal-to-noise ratio (SNR) regime. Moreover, this paper also investigates the optimality of the greedy policy for certain families of energy-arrival distributions, and establishes the tightest semi-universal bounds on the battery-capacity threshold for greedy optimality.

cs.IT↗

arXiv2026

Boolean Satisfiability via Imitation Learning

We propose ImitSAT, a branching policy for conflict-driven clause learning (CDCL) solvers based on imitation learning for the Boolean satisfiability problem (SAT). Unlike previous methods that predict instance-level signals to improve CDCL branching indirectly, or rely on reinforcement learning and insufficient CDCL information to enhance branching, ImitSAT learns from expert KeyTrace that collapses a full run into the sequence of surviving decisions. Replaying a KeyTrace on the same instance is nearly conflict-free, providing dense decision-level supervision and directly reducing propagations -- the dominant contributor to wall-clock time. This prefix-conditioned supervision enables ImitSAT to reproduce high-quality branches without exploration, yielding faster convergence, stable training, and seamless integration into CDCL. Extensive experiments demonstrate that ImitSAT reduces propagation counts and runtime, outperforming state-of-the-art learned approaches. We released the source code and trained model at https://github.com/zewei-Zhang/ImitSAT

cs.AI↗

arXiv2026

A Computation of Tamarkin-Tsygan Calculus

We compute the full Tamarkin-Tsygan calculus of a Koszul algebra whose global dimension exceeds the number of generators. Our results show that even for algebras possessing an economic presentation and agreeable homological properties, the Hochschild (co)homology, as well as the structure of the Tamarkin--Tsygan calculus may exhibit a rather intricate behavior.

math.RA↗

arXiv2026

Riemannian Momentum Tracking: Distributed Optimization with Momentum on Compact Submanifolds

Gradient descent with momentum has been widely applied in various signal processing and machine learning tasks, demonstrating a notable empirical advantage over standard gradient descent. However, momentum-based distributed Riemannian algorithms have been only scarcely explored. In this paper, we propose Riemannian Momentum Tracking (RMTracking), a decentralized optimization algorithm with momentum over a compact submanifold. Given the non-convex nature of compact submanifolds, the objective function, composed of a finite sum of smooth (possibly non-convex) local functions, is minimized across agents in an undirected and connected network graph. With a constant step-size, we establish an $\mathcal{O}(\frac{1-β}{K})$ convergence rate of the Riemannian gradient average for any momentum weight $β\in [0,1)$. Especially, RMTracking can achieve a convergence rate of $\mathcal{O}(\frac{1-β}{K})$ to a stationary point when the step-size is sufficiently small. To best of our knowledge, RMTracking is the first decentralized algorithm to achieve exact convergence that is $\frac{1}{1-β}$ times faster than other related algorithms. Finally, we verify these theoretical claims through numerical experiments on eigenvalue problems.

math.OC↗

arXiv2026

Robust Pedestrian Detection with Uncertain Modality

Existing cross-modal pedestrian detection (CMPD) employs complementary information from RGB and thermal-infrared (TIR) modalities to detect pedestrians in 24h-surveillance systems.RGB captures rich pedestrian details under daylight, while TIR excels at night. However, TIR focuses primarily on the person's silhouette, neglecting critical texture details essential for detection. While the near-infrared (NIR) captures texture under low-light conditions, which effectively alleviates performance issues of RGB and detail loss in TIR, thereby reducing missed detections. To this end, we construct a new Triplet RGB-NIR-TIR (TRNT) dataset, comprising 8,281 pixel-aligned image triplets, establishing a comprehensive foundation for algorithmic research. However, due to the variable nature of real-world scenarios, imaging devices may not always capture all three modalities simultaneously. This results in input data with unpredictable combinations of modal types, which challenge existing CMPD methods that fail to extract robust pedestrian information under arbitrary input combinations, leading to significant performance degradation. To address these challenges, we propose the Adaptive Uncertainty-aware Network (AUNet) for accurately discriminating modal availability and fully utilizing the available information under uncertain inputs. Specifically, we introduce Unified Modality Validation Refinement (UMVR), which includes an uncertainty-aware router to validate modal availability and a semantic refinement to ensure the reliability of information within the modality. Furthermore, we design a Modality-Aware Interaction (MAI) module to adaptively activate or deactivate its internal interaction mechanisms per UMVR output, enabling effective complementary information fusion from available modalities.

cs.CV↗

arXiv2026

Disorder-driven exceptional points and concurrent topological phase transitions in non-Hermitian systems

Exceptional points (EPs) are spectral degeneracies unique to non-Hermitian systems which underpin phenomena from enhanced sensing to unconventional topology. While disorder is usually viewed as detrimental, it can also drive topological phase transitions (TPTs). Here, we show that random disorder alone can generate EPs and concurrent TPTs in a multiorbital non-Hermitian lattice with nonreciprocal hopping. Increasing disorder induces successive real-complex-real spectral transitions accompanied by band inversion and quantized changes in the spin Bott index. Using effective medium theory and large-scale simulations, we trace these transitions to a competition between disorder-induced energy-level renormalization and nonreciprocity-driven hybridization. The resulting phase diagram reveals extended EP lines that emerge from the Hermitian TPT point and persist over a broad parameter range. Our results establish disorder as an active mechanism for engineering exceptional point mediated topology in non-Hermitian matter.

cond-mat.dis-nn↗

arXiv2026

Three-dimensional properties of a coronal shock and the longitudinal distribution of its related solar energetic particles

This study aims to investigate the relationship between the spatial-temporal evolution of shock properties and the longitudinal dependence of SEP intensities and spectra. The shock parameters, including the normal speed, oblique angles, compression ratio, and Alfven Mach number, were derived by combining a steady-state solar-wind simulation with the three-dimensional (3D) reconstruction of the shock surface based on multi-view observations. We compared the local shock parameters at the magnetic connecting points with in situ proton intensities and peak spectra to establish the link between shock evolution and SEP characteristics. The shock nose consistently exhibited higher particle-acceleration efficiency with the largest normal speed, compression ratio, and supercritical Alfven Mach number, while the flanks showed delayed transition to supercritical Alfven Mach number with weaker efficiency. The earliest and most rapid proton enhancement of STEREO-B correlated with efficient shock acceleration and prompt magnetic connectivity to the shock. Spectral analysis revealed that proton energy spectra were consistent with the relativistic diffusive shock acceleration (DSA) estimations. The initial shock acceleration began at about 1.4-5 Rsun and caused the widespread longitudinal SEP distribution. The longitudinal dependence of SEP intensity and spectral variations arise from the combined influence of 3D shock properties, magnetic connectivity, and particle transport processes. The agreement between in situ proton indices and relativistic DSA estimations supports DSA in this SEP event and provides insights into the early-stage acceleration at the source region.

astro-ph.SR↗

arXiv2026

Unexpected type-II multiferroic phase in GdMnO3 under high magnetic fields

Perovskite manganites with small A-site ions, as the first and canonical branch of type-II multiferroics, are ideal systems to exhibit magnetism-induced ferroelectricity. Despite their established magnetoelectric phase diagrams under low magnetic fields, here an unidentified phase with a large magnetism-induced polarization (up to 1500 μC/m2) is revealed in GdMnO3 under high magnetic fields up to 60 T. Based on multiprobe experiments, a complete phase diagram is constructed with successive polar-nonpolar-polar-nonpolar transitions. Such a nonmonotonic evolution is well mimicked by model simulation, while the spin-lattice coupling is the key ingredient for the reentrant ferroelectric phase.

cond-mat.mtrl-sci↗

arXiv2026

Magnetoelectric torque in polar magnetic bilayers

Energy-efficient fast switching of spin orientations or textures is a core issue of spintronics, which is highly demanded but remains challenging. Different from the mainstream routes based on spin-transfer torque or spin-orbit torque, here we propose another mechanism coined as magnetoelectric torque to switch the magnetization in polar magnetic bilayers via pure electric field. In some magnetic van der Waals bilayers, when the electrostatic energy of polarization can compensate the interlayer magnetic coupling, a magnetoelectric torque is generated to fastly flip spins within a few picoseconds, which is demonstrated by combining the first-principles calculations, analytic model, as well as atomistic simulations. Such a magnetoelectric torque doesn't rely on the spin-orbit coupling and is generally active in polar magnetic homostructures and heterostructures. Our work provides an alternative route to switch magnetization in nanoscale, which may benefit the energy-saving and fast-response spintronic devices.

cond-mat.mtrl-sci↗

arXiv2026

Modulating near-field radiative energy and momentum transfer via rotating Weyl semimetals

We study near-field radiative transfer of energy, angular momentum, and linear momentum between a nanoparticle and a plate consisting of magnetic Weyl semimetals, and demonstrate that these can be efficiently tuned by a relative angle between the Weyl node separations. This tunability originates from the coupling between the particle-induced rotational Poynting vector and the nonreciprocal surface plasmon polaritons supported by the plate. Remarkably, we uncover a counterintuitive regime in which both energy and angular momentum transfer are maximized when the Weyl node separations are antiparallel rather than parallel. This arises from optimal mode matching between the rotation direction of the particle's circular heat flux and the propagation direction of the surface plasmon polaritons in the antiparallel configuration.

physics.optics↗

arXiv2025

Modeling and Analysis of Land-to-Ship Maritime Wireless Channels at 5.8 GHz

Maritime channel modeling is crucial for designing robust nearshore communication systems, yet reliable models that account for the dynamic marine environment with varying sea waves, wind conditions, and vessel motions remain scarce. This article investigates land-to-ship maritime wireless channel characteristics at 5.8 GHz based upon an extensive measurement campaign, with concurrent hydrological and meteorological information collection. First, a novel large-scale path loss model with physical foundation and high accuracy is proposed for dynamic marine environments. Then, we introduce the concept of sea-wave-induced fixed-point (SWIFT) fading, a peculiar phenomenon in maritime scenarios that captures the impact of sea surface fluctuations on received power. An enhanced two-ray model incorporating vessel rotational motion is propounded to simulate the SWIFT fading, showing good alignment with measured data, particularly for modest antenna movements. Next, the small-scale fading is studied by leveraging a variety of models including the two-wave with diffuse power (TWDP) and asymmetric Laplace distributions, with the latter performing well in most cases, while TWDP better captures bimodal fading in rough seas. Furthermore, maritime channel sparsity is examined via the Gini index and Rician $K$ factor, and temporal dispersion is characterized. The resulting channel models and parameter characteristics offer valuable insights for maritime wireless system design and deployment.

eess.SP↗

arXiv2025

Fast, accurate measurement of the worker populations of honey bee colonies using deep learning

Honey bees play a crucial role in pollination, contributing significantly to global agriculture and ecosystems. Accurately estimating hive populations is essential for understanding the effects of environmental factors on bee colonies, yet traditional methods of counting bees are time-consuming, labor-intensive, and prone to human error, particularly in large-scale studies. In this paper, we present a deep learning-based solution for automating bee population counting using CSRNet and introduce ASUBEE, the FIRST high-resolution dataset specifically designed for this task. Our method employs density map estimation to predict bee populations, effectively addressing challenges such as occlusion and overlapping bees that are common in hive monitoring. We demonstrate that CSRNet achieves superior performance in terms of time efficiency, with a computation time of just 1 second per image, while delivering accurate counts even in complex and densely populated hive scenarios. Our findings show that deep learning approaches like CSRNet can dramatically enhance the efficiency of hive population assessments, providing a valuable tool for researchers and beekeepers alike. This work marks a significant advancement in applying AI technologies to ecological research, offering scalable and precise monitoring solutions for honey bee populations.

q-bio.QM↗

arXiv2025

Step-Audio-R1 Technical Report

Recent advances in reasoning models have demonstrated remarkable success in text and vision domains through extended chain-of-thought deliberation. However, a perplexing phenomenon persists in audio language models: they consistently perform better with minimal or no reasoning, raising a fundamental question - can audio intelligence truly benefit from deliberate thinking? We introduce Step-Audio-R1, the first audio reasoning model that successfully unlocks reasoning capabilities in the audio domain. Through our proposed Modality-Grounded Reasoning Distillation (MGRD) framework, Step-Audio-R1 learns to generate audio-relevant reasoning chains that genuinely ground themselves in acoustic features rather than hallucinating disconnected deliberations. Our model exhibits strong audio reasoning capabilities, surpassing Gemini 2.5 Pro and achieving performance comparable to the state-of-the-art Gemini 3 Pro across comprehensive audio understanding and reasoning benchmarks spanning speech, environmental sounds, and music. These results demonstrate that reasoning is a transferable capability across modalities when appropriately anchored, transforming extended deliberation from a liability into a powerful asset for audio intelligence. By establishing the first successful audio reasoning model, Step-Audio-R1 opens new pathways toward building truly multimodal reasoning systems that think deeply across all sensory modalities.

cs.AI↗