arXiv subjects

Jianing Li

Publications and source records attributed to Jianing Li.

At least 19 recordsLinked to original sources

arXiv2026

Unveiling QCD Criticality with Cross-Rapidity Net-Baryon Cumulants

Fluctuations of conserved charges are a primary tool in the search for the QCD critical endpoint, but their beam-energy dependence is complicated by global baryon-number conservation, whose influence changes as the experimental acceptance covers different fractions of the collision system. We propose using correlations of net-baryon fluctuations between two separated rapidity windows to exploit the distinct signatures of conservation and critical dynamics. Global conservation produces a negative cross-window correlation, whereas a common long-wavelength critical fluctuation produces a positive one. We show that, within a canonical independent-source framework, the conservation-induced background and the critical signal enter additively at leading order, allowing the leading conservation term to be estimated and subtracted. The resulting yield-scaled correlator follows the nonmonotonic enhancement of an Ising-mapped equilibrium correlation length along a freeze-out trajectory passing near a hypothetical critical endpoint, while wider rapidity windows reduce the response through thermal smearing. Cross-rapidity cumulants therefore provide a rapidity-differential strategy for reducing the leading conservation background and sharpening fluctuation-based searches for QCD criticality in beam-energy-scan experiments.

nucl-th↗

arXiv2026

Hadronic rescattering effects on net-proton cumulants from functional renormalization group calculations

Net-proton cumulants in the Beam Energy Scan region of heavy-ion collisions are widely used to probe critical fluctuations associated with the conjectured critical endpoint of Quantum Chromodynamics (QCD). Most existing studies, however, concentrate on the initial-state or phase-transition contributions, while the impact of hadronic rescattering on these observables has not been fully quantified. To address this gap, we construct event-by-event proton and antiproton distributions from functional renormalization group (fRG) cumulants using the maximum entropy principle, and propagate the resulting particles through the hadronic transport model SMASH in a simplified spherical evolution setup. We systematically investigate how the hadronic cascade modifies net-proton cumulants at collision energies $\sqrt{s_{NN}}=3.0$, 3.9, 4.9, 7.2, and 7.7~GeV. In the canonical-ensemble framework, which enforces exact net-baryon number conservation, the higher-order cumulant signal---in particular the ratio $C_4/C_2$ at $\sqrt{s_{NN}}=4.9$~GeV---is strongly reduced during the early stage of the cascade; the suppression of $C_4/C_2$ reaches approximately $20\%$. The non-monotonic energy dependence inherited from the fRG input survives the hadronic evolution, but its magnitude is substantially modified. These results demonstrate that hadronic rescattering provides a non-negligible background effect that must be accounted for when extracting QCD critical-point signals from experimental data.

nucl-th↗

arXiv2026

Magic-wavelength matter-wave interferometry with optical clock states

Optical clocks and atom interferometers provide complementary ways to measure time, motion and gravity. Combining these capabilities requires matter-wave beam splitters that manipulate different clock states in the same way, so that optical internal energy becomes a controlled degree of freedom rather than a source of systematic phase shifts. Here we realized a dual matter-wave interferometer operating simultaneously on the two states of the $^{88}$Sr optical clock transition, $^1S_0$ and $^3P_0$. The interferometer is driven by Bragg pulses at the 813 nm magic wavelength, for which the two clock states experience the same optical coupling strength. This realizes a common matter-wave beam splitter for atoms whose internal energies differ by an optical excitation. With a sensitivity of 30 mrad, our measurement is consistent with a zero differential phase shift between the two clock-state Mach-Zehnder interferometers, translating to an absence of state-dependent acceleration in free fall at the level of $10^{-5}$. We further used the same interferometer to measure state-dependent optical dipole forces and determine a tune-out wavelength of the metastable $^3P_0$ state to be 478.95(8) nm. These results establish magic-wavelength clock-state interferometry as a platform for differential force sensing, excited-state polarizability metrology and future quantum-clock tests of gravity.

physics.atom-ph↗

arXiv2026

Can LVLMs Uncover the Truth Behind Visual Illusions? An Analysis of Perceptual and Reasoning Capabilities

Large Vision Language Models have integrated reasoning capabilities, elevating cognitive performance to new levels. However, existing evaluations either focus solely on perception or rely on specific domains such as maths or coding. Evaluation for reasoning capabilities that align with an open-world environment is still required, especially one that considers perception and reasoning jointly. To bridge this gap, we propose to evaluate LVLMs by exploiting visual illusions as a diagnostic tool. Visual illusions are phenomena in which the human visual system misinterprets objective signals, resulting in an understanding that deviates from reality. We constructed Illusion-Reasoning, a benchmark of illusion images collected from the real world, incorporating diverse annotated question-answer pairs. Based on Illusion-Reasoning, we show that the reasoning capabilities of a wide range of LVLMs are not as advanced as claimed. Our work provides new insights into LVLMs and offers future directions for optimisation. Our project is publicly available at https://github.com/zhaoliangjie55/EMNLP2026_Illusion.

cs.CL↗

arXiv2026

Existence of generalized bent functions in the exceptional $q\equiv2\pmod4$, odd-dimensional case

We resolve an open problem of Kumar, Scholtz, and Welch (1985) by constructing generalized bent functions from $(\mathbb{Z}/q\mathbb{Z})^m$ to $\mathbb{Z}/q\mathbb{Z}$ in the exceptional case $q\equiv2\pmod4$ with $m$ odd, the case their paper left without a construction and which four decades of subsequent work had addressed only through nonexistence results. Concretely, for all integers $m\geq d\geq 2$ with $m=dr+2k$, $r\geq 1$, $k\geq 0$, we construct an explicit generalized bent function from $(\mathbb{Z}/q\mathbb{Z})^m$ to $\mathbb{Z}/q\mathbb{Z}$, where $q=2(2^d-1)$. We further show that, when $d\ge3$, these generalized bent functions have Fourier coefficients that are not roots of unity --- all of them when $r$ is odd --- which gives a negative answer to a recent question of Armario, Egan, Kharaghani, and \'O~Cath\'ain about bent vectors for character tables.

math.CO↗

arXiv2026

Neuromorphic Object Detection: An In-Depth Study and Future Directions

Conventional frame-based cameras face significant challenges in detecting objects under high-speed motion blur or in low-light environments. Neuromorphic cameras provide asynchronous visual streams with high temporal resolution and a wide dynamic range, offering a promising solution for object detection under challenging conditions. Despite the development of numerous models and the emergence of various applications in neuromorphic object detection, there is still a lack of deep understanding and standardized benchmarks to assess progress and address key challenges. In this paper, we provide a comprehensive survey and benchmark of existing neuromorphic object detection algorithms. Specifically, we first present a problem description, review the available datasets, and revisit the evaluation metrics. We then explore existing neuromorphic object detection approaches from various perspectives, including event representation, temporal modeling, multimodal fusion, asynchronous processing, low-latency processing, and energy-efficient computing. Furthermore, we evaluate a wide range of representative neuromorphic object detection models and offer detailed analyses of the comparative results. Finally, we discuss unresolved issues in neuromorphic object detection and propose potential future research directions. We hope this survey and benchmark will be a valuable resource for researchers and provide guidance for future advancements in neuromorphic object detection.

cs.CV↗

arXiv2026

Towards Ultrafast Depth Sensing Via Active Event-based Stereo Vision

Conventional frame-based imaging for active stereo systems has encountered major challenges in fast-motion scenarios. However, how to design a novel paradigm for ultrafast depth sensing remains an open issue. In this paper, we propose a novel problem setting, namely active event-based stereo vision, which attempts to integrate binocular event cameras and an infrared 2D pattern projector for high-speed dense depth sensing. Technically, we first build a stereo camera prototype system and present a real-world dataset with over 21.5k spatiotemporal synchronized labels at 15 Hz, while also establishing a realistic synthetic dataset with stereo event streams and 23.8k synchronized labels at 20 Hz. Then, we propose ActiveEventNet+, a lightweight yet effective event-based stereo matching neural network that learns to generate high-quality dense disparity maps from stereo event streams with low latency. Our ActiveEventNet+ mainly involves three innovations: incorporating lightweight blocks into event-based stereo matching frameworks, designing a novel cost volume with dynamic interactions between stereo pairs, and presenting an effective temporal consistency architecture to fully use rich temporal cues in event streams. The results show that our ActiveEventNet+ outperforms state-of-the-art methods while significantly reducing computational complexity. Our solution offers superior depth sensing performance compared to conventional frame-based stereo cameras in high-speed scenes. In particular, the lightweight ActiveEventNet enables the prototype system to achieve real-time processing at speeds up to 150 FPS. We believe that this novel active event-based stereo vision paradigm can provide new insights into the design of future high-speed depth sensing camera systems. Our dataset and code can be available at https://github.com/jianing-li/active_event_based_stereo.

cs.RO↗

arXiv2026

Explaining and Tuning Transformer-based LLMs in Arithmetic Tasks with Human Strategies

Transformer-based large language models (LLMs) continue to achieve state-of-the-art performance across various natural language processing tasks. However, their subpar performance on seemingly elementary problems, such as basic arithmetic, raises concerns about model reliability, safety, and ethical deployment. In this study, we demonstrate that the performance of a vanilla Transformer model trained on integer arithmetic tasks can be improved using methods effective for human learners. We begin by decomposing the arithmetic task into well-defined subtasks and conducting loss convergence order analysis together with ablation studies for each subtask. Our findings reveal that LLMs exhibit learning patterns similar to those of human learners, with a faster learning speed for simpler subtasks compared to more complex ones. In addition, we successfully improved the accuracy of LLMs by applying problem-solving strategies and cognitive empowerment methods shown to enhance the performance of human learners. This suggests that transformer-based LLMs may share cognitive processes with human learners in arithmetic. Lastly, we provide a comprehensive demonstration of our method's effectiveness, including significant accuracy improvement experiments, visualization verification, and explanation-based analysis to illuminate the intricacies of LLMs in arithmetic learning. In general, this work explores the potential similarities between transformer-based LLMs and human learners, supported by explainable AI (XAI) verifications, ultimately fostering trust in LLMs for critical and high-stakes applications.

cs.LG↗

arXiv2026

Occ-VLM: Occupancy Grounded Vision Language Model for Indoor Scene Understanding

Recently, vision-language models (VLMs) have made significant progress in 3D scene understanding, driving advances in applications such as embodied intelligence and robotic vision. However, existing approaches typically either rely directly on explicit 3D inputs (e.g., point clouds or RGB-D sequences), or introduce an additional 3D geometry encoder to derive 3D-aware visual tokens from 2D images. Such designs structurally decouple 3D geometric perception from the rich 2D semantics learned via vision-language pre-training, hindering the development of a unified 3D vision-language representation. In this work, we propose Occ-VLM, a novel framework for 3D scene understanding that operates purely on posed RGB images and employs a single 2D vision encoder. Specifically, Occ-VLM reconstructs 3D scene occupancy as an auxiliary geometric prior, which is utilized to spatially associate foreground 2D tokens with 3D space. These tokens are then decoded by a Large Language Model (LLM) for unified scene understanding. Extensive experiments demonstrate that Occ-VLM achieves both accurate geometric perception and robust vision-language reasoning: it attains state-of-the-art performance on multi-view occupancy prediction, while performing on par with 3D-input VLMs on 3D Visual Question Answering (VQA) and 3D dense captioning benchmarks.

cs.CV↗

arXiv2026

Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3

Antimicrobial resistance causes to over a million deaths annually. Antimicrobial peptides (AMPs) are a promising solution, but generative AMP models are not yet ready to design peptides with non-natural amino acids and/or chemical modifications, which are essential for real-world peptide drugs. We present AMPGAN v3, a multi-objective conditional GAN that expands the generative vocabulary to D-amino acids and N/C-terminus modifications such as amidation. By separating adversarial and activity-aware supervision across two specialized discriminators, AMPGAN v3 substantially improves training stability and outperforms prior generative AMP models on external classifiers. We validated five candidates spanning three structural classes in vitro; two showed activity against Gram-positive strains, with the best candidate reaching MIC 8 {\mu}g/mL against B. subtilis. To support downstream curation, we further present PepCraft, a multi-agent framework for end-to-end AMP discovery in which a Planning Agent orchestrates specialized executors for generation, filtering, and verification. Its prioritization recommendations align with our in vitro outcomes. Together, these contributions let us examine, on a small but real scale, how generative and agentic AI compose in therapeutic peptide discovery. Code: https://github.com/marszzibros/AMPGANv3

q-bio.QM↗

arXiv2026

Looping Back to Move Forward: Recursive Transformers for Efficient and Flexible Large Multimodal Models

Large Multimodal Models (LMMs) have achieved remarkable success in vision-language tasks, yet their vast parameter counts are often underutilized during both training and inference. In this work, we embrace the idea of looping back to move forward: reusing model parameters through recursive refinement to extract stronger multimodal representations without increasing model size. We propose RecursiveVLM, a recursive Transformer architecture tailored for LMMs. Two key innovations enable effective looping: (i) a Recursive Connector that aligns features across recursion steps by fusing intermediate-layer hidden states and applying modality-specific projections, respecting the distinct statistical structures of vision and language tokens; (ii) a Monotonic Recursion Loss that supervises every step and guarantees performance improves monotonically with recursion depth. This design transforms recursion into an on-demand refinement mechanism: delivering strong results with few loops on resource-constrained devices and progressively improving outputs when more computation resources are available. Experiments show consistent gains of +3% over standard Transformers and +7% over vanilla recursive baselines, demonstrating that strategic looping is a powerful path toward efficient, deployment-adaptive LMMs.

cs.LG↗

arXiv2026

Dual Tuning for Reasoning Efficacy-Driven Data Curation in Multimodal LLM Training

Reasoning post-training improves Large Language Models (LLMs) on complex tasks such as mathematics and coding, but its benefits across diverse multimodal tasks remains uncertain. The trend of releasing parallel "Instruct" and "Thinking" models by leading teams is both resource-intensive and user-unfriendly. Prior work finds that the gains from reasoning training are influenced by multiple factors, such as base model capabilities, task characteristics, and Chain-of-Thought (CoT) data quality. However, principled criteria for determining when reasoning post-training is beneficial and which data should support it are still lacking. In this paper, we propose Dual Tuning, a reasoning efficacy-driven data curation framework for multimodal LLMs training. Given a target task and a base model, Dual Tuning jointly evaluates whether the training data is beneficial and whether reasoning training with current CoT content yields positive gains over non-reasoning alternatives. We apply Dual Tuning across spatial, mathematical, and multi-disciplinary tasks, and further analyze how reinforcement learning and thinking patterns affect reasoning efficacy. The Dual Tuning results guide data curation by identifying data that benefit reasoning training, data better suited to direct-answer training, and data that are detrimental under both training modes. Our work provides quantitative criteria for selecting appropriate training data and matching post-training strategies.

cs.CL↗

arXiv2026

EventFlash: Towards Efficient MLLMs for Event-Based Vision

Event-based multimodal large language models (MLLMs) enable robust perception in high-speed and low-light scenarios, addressing key limitations of frame-based MLLMs. However, current event-based MLLMs often rely on dense image-like processing paradigms, overlooking the spatiotemporal sparsity of event streams and resulting in high computational cost. In this paper, we propose EventFlash, a novel and efficient MLLM to explore spatiotemporal token sparsification for reducing data redundancy and accelerating inference. Technically, we build EventMind, a large-scale and scene-diverse dataset with over 500k instruction sets, providing both short and long event stream sequences to support our curriculum training strategy. We then present an adaptive temporal window aggregation module for efficient temporal sampling, which adaptively compresses temporal tokens while retaining key temporal cues. Finally, a sparse density-guided attention module is designed to improve spatial token efficiency by selecting informative regions and suppressing empty or sparse areas. Experimental results show that EventFlash achieves a $12.4\times$ throughput improvement over the baseline (EventFlash-Zero) while maintaining comparable performance. It supports long-range event stream processing with up to 1,000 bins, significantly outperforming the 5-bin limit of EventGPT. We believe EventFlash serves as an efficient foundation model for event-based vision.

cs.CV↗

arXiv2026

Resource-Efficient Reinforcement for Reasoning Large Language Models via Dynamic One-Shot Policy Refinement

Large language models (LLMs) have exhibited remarkable performance on complex reasoning tasks, with reinforcement learning under verifiable rewards (RLVR) emerging as a principled framework for aligning model behavior with reasoning chains. Despite its promise, RLVR remains prohibitively resource-intensive, requiring extensive reward signals and incurring substantial rollout costs during training. In this work, we revisit the fundamental question of data and compute efficiency in RLVR. We first establish a theoretical lower bound on the sample complexity required to unlock reasoning capabilities, and empirically validate that strong performance can be achieved with a surprisingly small number of training instances. To tackle the computational burden, we propose Dynamic One-Shot Policy Refinement (DoPR), an uncertainty-aware RL strategy that dynamically selects a single informative training sample per batch for policy updates, guided by reward volatility and exploration-driven acquisition. DoPR reduces rollout overhead by nearly an order of magnitude while preserving competitive reasoning accuracy, offering a scalable and resource-efficient solution for LLM post-training. This approach offers a practical path toward more efficient and accessible RL-based training for reasoning-intensive LLM applications.

cs.AI↗

arXiv2026

NFCDS: A Plug-and-Play Noise Frequency-Controlled Diffusion Sampling Strategy for Image Restoration

Diffusion sampling-based Plug-and-Play (PnP) methods produce images with high perceptual quality but often suffer from reduced data fidelity, primarily due to the noise introduced during reverse diffusion. To address this trade-off, we propose Noise Frequency-Controlled Diffusion Sampling (NFCDS), a spectral modulation mechanism for reverse diffusion noise. We show that the fidelity-perception conflict can be fundamentally understood through noise frequency: low-frequency components induce blur and degrade fidelity, while high-frequency components drive detail generation. Based on this insight, we design a Fourier-domain filter that progressively suppresses low-frequency noise and preserves high-frequency content. This controlled refinement injects a data-consistency prior directly into sampling, enabling fast convergence to results that are both high-fidelity and perceptually convincing--without additional training. As a PnP module, NFCDS seamlessly integrates into existing diffusion-based restoration frameworks and improves the fidelity-perception balance across diverse zero-shot tasks.

cs.CV↗

arXiv2026

Principal ideal problem and ideal shortest vector over rational primes in power-of-two cyclotomic fields

The shortest vector problem (SVP) over ideal lattices is closely related to the Ring-LWE problem, which is widely used to build post-quantum cryptosystems. Power-of-two cyclotomic fields are frequently adopted to instantiate Ring-LWE. Pan et al. (EUROCRYPT~2021) explored the SVP over ideal lattices via the decomposition fields and, in particular determined the length of the shortest vector in prime ideals lying over rational primes $p\equiv3,5\pmod{8}$ in power-of-two cyclotomic fields via explicit construction of reduced lattice bases. In this work, we first provide a new method (different from analyzing lattice bases) to analyze the length of the shortest vector in prime ideals in $\mathbb{Z}[\zeta_{2^{n+1}}]$ when $p\equiv3,5\pmod{8}$. Then we precisely characterize the length of the shortest vector in the cases of $p\equiv7,9\pmod{16}$. Furthermore, we derive a new upper bound $\sqrt[4]{2^{2n+1}p}$ for this length, which is tighter than the bound $2^n\sqrt[4]{p}$ obtained from Minkowski's theorem. Our key technique is to investigate whether a generator of a principal ideal can achieve the shortest length after embedding as a vector. If this holds for the ideal, finding the shortest vector in this ideal can be reduced to finding its shortest generator.

cs.CR↗

arXiv2025

Learning to Remove Lens Flare in Event Camera

Event cameras have the potential to revolutionize vision systems with their high temporal resolution and dynamic range, yet they remain susceptible to lens flare, a fundamental optical artifact that causes severe degradation. In event streams, this optical artifact forms a complex, spatio-temporal distortion that has been largely overlooked. We present E-Deflare, the first systematic framework for removing lens flare from event camera data. We first establish the theoretical foundation by deriving a physics-grounded forward model of the non-linear suppression mechanism. This insight enables the creation of the E-Deflare Benchmark, a comprehensive resource featuring a large-scale simulated training set, E-Flare-2.7K, and the first-ever paired real-world test set, E-Flare-R, captured by our novel optical system. Empowered by this benchmark, we design E-DeflareNet, which achieves state-of-the-art restoration performance. Extensive experiments validate our approach and demonstrate clear benefits for downstream tasks. Code and datasets are publicly available.

cs.CV↗

arXiv2025

Semantic Temporal Single-photon LiDAR

Temporal single-photon (TSP-) LiDAR presents a promising solution for imaging-free target recognition over long distances with reduced size, cost, and power consumption. However, existing TSP-LiDAR approaches are ineffective in handling open-set scenarios where unknown targets emerge, and they suffer significant performance degradation under low signal-to-noise ratio (SNR) and short acquisition times (fewer photons). Here, inspired by semantic communication, we propose a semantic TSP-LiDAR based on a self-updating semantic knowledge base (SKB), in which the target recognition processing of TSP-LiDAR is formulated as a semantic communication. The results, both simulation and experiment, demonstrate that our approach surpasses conventional methods, particularly under challenging conditions of low SNR and limited acquisition time. More importantly, our self-updating SKB mechanism can dynamically update the semantic features of newly encountered targets in the SKB, enabling continuous adaptation without the need for extensive retraining of the neural network. In fact, a recognition accuracy of 89% is achieved on nine types of unknown targets in real-world experiments, compared to 66% without the updating mechanism. These findings highlight the potential of our framework for adaptive and robust target recognition in complex and dynamic environments.

eess.IV↗