Searcharxiv⌕ Search

arXiv subjects

Peng Liu

Publications and source records attributed to Peng Liu.

At least 19 recordsLinked to original sources
arXiv2026

VLX-VR: An Agentic-Aware Video Reasoning Model

Real-world video understanding requires integrating visual, audio, textual, and temporal evidence distributed across a video. Yet many pipelines use a fixed video context and single-pass inference, limiting adaptive evidence acquisition when observations are incomplete, ambiguous, or conflicting. We present VLX-VR, an agentic-aware video reasoning model trained within a video reasoning framework defined by a Think--Memory--Observation loop. At each step, VLX-VR determines the needed evidence, invokes read_memory or write_memory, incorporates the returned Observation, and decides whether to continue or produce the task output. We train VLX-VR with multimodal data, including videos and agent trajectories, using reinforcement learning to learn evidence acquisition, memory use, and termination. On MINERVA, VLX-VR achieves state-of-the-art performance among the models included in our comparison, with 78.79% accuracy. Under the original three duration groups, its accuracies are 76.70%, 78.73%, and 80.92%, with a cross-duration accuracy variance of 2.97~$\mathrm{pp}^2$. On correctly answered samples, 96.20% of VLX-VR's reasoning traces are consistent with the MINERVA reference reasoning traces and the evidence described by them, while approximately 75.80% of all evaluated samples satisfy both answer correctness and this evidence-grounded trace criterion. These results show strong performance and broadly stable behavior across durations, while counting, state changes, causal reasoning, and spatial perception remain challenging.

cs.CL↗
arXiv2026

Safe Task Planning with Long-Term Graph Memory for Embodied Agents

Large language models (LLMs) and vision-language models (VLMs) have significantly advanced zero-shot task planning for embodied agents. However, most LLM- and VLM-driven methods struggle to generate safe high-level actions due to a lack of physical risk awareness, particularly under partial observability, where hazards lie outside the immediate field of view. To address this challenge, we propose a novel safe task-planning framework, SafeMem, which constructs and maintains a long-term semantic graph memory of the open and dynamic environment. Based on egocentric observations, the proposed framework incrementally accumulates knowledge about surrounding objects and their relationships with a graph. Then, an LLM-based risk predictor evaluates candidate actions using the graph memory, triggering a conservatism-modulated replanning loop with explanations for detected hazards. Extensive experiments on the IS-Bench benchmark and a real-world robot platform demonstrate that the SafeMem framework substantially improves safe success rates compared to state-of-the-art VLM-driven task planners. Video results are available on our webpage: https://sites.google.com/view/safemem.

cs.RO↗
arXiv2026

Portfolio Diversification and Concentration under Dependence Uncertainty: A Majorization Approach

Modern portfolio theory identifies diversification as the primary tool for risk reduction. However, under model uncertainty, this cornerstone may no longer remain optimal. This paper investigates the tension between portfolio diversification and concentration under dependence uncertainty. In the absence of model uncertainty, we employ the framework of the majorization order and doubly stochastic matrices to formalize the degree of diversification, and prove that quasi-convexity is a necessary and sufficient property for a risk functional to be weakly consistent with the majorization order. We further derive worst-case risk measure inequalities and solve robust portfolio selection problems for a broad class of risk measures, including VaR, ES, Range-VaR (RVaR), and standard deviation (SD). Our results reveal a ''concentration paradox'' for many widely-used risk functionals: when the dependence structure is fully ambiguous, robust optimization often recommends concentrating investment in a single asset to hedge against the worst-case dependence scenario. As an application, we propose a weighted robustness formulation that interpolates between a reference dependence structure and the worst-case structure. The formulation is structurally analogous to the constrained/unconstrained Expected Shortfall blend in the Fundamental Review of the Trading Book (FRTB) and provides a theoretical foundation for balancing diversification against robustness in the presence of model uncertainty.

q-fin.PM↗
arXiv2026

Moirae: A Multimodal Agent Collaborative Framework for Dynamic Android Malware Detection

The Android ecosystem faces persistent and rapidly evolving malware threats. Existing machine learning detectors are vulnerable to concept drift because they rely on implementation-specific features whose distributions change over time. Large language models (LLMs) offer strong semantic understanding and zero-shot reasoning, but current LLM-based detectors typically depend on code-centric or single-dimensional evidence, making them susceptible to obfuscation and limiting comprehensive behavior analysis. We present {\sysname}, a multimodal agent collaborative framework for dynamic Android malware detection. {\sysname} dynamically collects multimodal runtime evidence and employs ReAct-based specialized agents to analyze complementary behavioral views. The detection process begins by identifying visual deception cues, modeling UI state transitions, and integrating runtime API behaviors to fuse multi-dimensional evidence across user-visible interfaces and hidden backend operations. Experiments on temporally and distributionally unseen datasets show that {\sysname} achieves an accuracy of 90.06\% without fine-tuning, outperforming state-of-the-art baselines and demonstrating strong zero-shot generalization against Android malware concept drift.

cs.CR↗
arXiv2026

Resource-Efficient Pruning for Transformer via Low-Rank Importance Estimation

With the rapid development of large-scale pre-trained language models based on Transformer architectures, their high computational and memory costs have become a major obstacle to deployment, especially in resource-constrained environments. Traditional pruning methods typically depend on full gradient-based importance estimation, and they necessitate prior finetuning of the model to achieve satisfactory performance. This process often results in intolerable resource consumption. This paper proposes REP-LIE, a new approach to enable resource-efficient pruning during the process of finetuning. REP-LIE leverages the gradients of LoRA low-rank matrices to estimate the importance of weights without requiring full gradient computation. To address the inherent randomness in importance estimation, a stability score is introduced, serving as the basis for iterative pruning of unimportant model parameters. The pruned model is further finetuned through lightweight updates, eliminating the need for full-parameter optimization in the process of finetuning. Extensive experiments on both medium-scale encoder models and large-scale generative models (LLaMA-7B and Mistral-7B) demonstrate that REP-LIE still achieves competitive performance compared to existing approaches.

cs.LG↗
arXiv2026

Discovery of Three Glitches in the previously quiet pulsar PSR J1637$-$4642

We present the discovery and analysis of three rotational glitches in the young pulsar PSR J1637$-$4642. The timing observations span from 19 February 2009 to 6 October 2024 (MJD 54881$-$60589) from the Murriyang radio telescope of the Parkes Observatory. The first and strongest glitch occurred around MJD 58352 with a fractional frequency change of $\Delta\nu/\nu \sim 2.7 \times 10^{-6}$, while two additional smaller glitches were detected at MJD 59443 and MJD 60445 with fractional changes of $2.2 \times 10^{-9}$ and $2.8 \times 10^{-8}$, respectively. Prior to this, the pulsar had shown no glitch activity since its discovery in the Parkes Multibeam survey. Only the first glitch exhibits detectable exponential recovery, with a decay timescale of $\sim$100 days and a small recovery fraction $\approx 0.015$, accompanied by a permanent increase in the magnitude of the spin-down rate. Modeling the post-glitch evolution of $\dot{\nu}$ within the vortex-creep framework using Bayesian inference gives a superfluid moment-of-inertia fraction $\approx 0.0187$, consistent with the inner-crust superfluid. These results reinforce the standard superfluid glitch paradigm and demonstrate that even ``quiet'' pulsars can still host substantial glitch activity.

astro-ph.HE↗
arXiv2026

Booster-based beam recycling for swap-out injection at the High Energy Photon Source

Fourth-generation synchrotron light sources employ ultralow-emittance storage rings with stringent injection requirements. On-axis swap-out injection alleviates the dependence on storage-ring dynamic aperture, but high-charge operation requires an efficient injector architecture capable of producing high-charge replacement bunches. This paper presents the accelerator physics design and performance analysis of a booster-based beam-recycling swap-out injection scheme implemented at the High Energy Photon Source (HEPS). In this approach, the full-energy booster serves as both an injector and a high-energy accumulator. An extracted storage-ring bunch is returned to the booster, merged with a low-charge bunch previously injected from the linac and accelerated to full energy. Following high-energy damping, the merged bunch is reinjected into the original storage-ring bucket. The scheme avoids the need for a dedicated accumulator ring while enabling high-charge bunch replacement. The recycling scheme was commissioned through staged machine studies. Full recycling-chain simulations, commissioning studies, and measured performance analysis are presented. The measured results characterize the recycling operation and quantify the transmission efficiency and performance limitations of the complete recycling loop. These results demonstrate the feasibility of the booster-based beam-recycling architecture and establish its operational basis for high-charge swap-out injection in future fourth-generation synchrotron light sources.

physics.acc-ph↗
arXiv2026

Sublime Transfer Printing of Three-Dimensional Nanostructure Ensembles

High-resolution three-dimensional (3D) nanostructures for visible-light photon manipulation provide unique and bespoke capabilities in optics and photonics. However subwavelength nanofabrication and reliable ensemble manipulation of the 3D prints onto arbitrary substrates remain challenging. Here, we introduce sublime transfer strategy tailored for transfer printing ensembles of delicate 3D printed nanostructures. This strategy enables conformal, damage-free integration of arrays of 3D structures on diverse substrates. Naphthalene acts as a transient stamp to encapsulate the structures during transfer and placement. We rely on the low sublimation temperature of naphthalene to release the structures reliably with nearly zero stress, preventing mechanical damage and positional misalignment. This approach is broadly applicable to integrate diverse nanostructures and photonic devices onto various substrates, and enabling inorganic architectures through ensemble uniform post-processing, including 2.5D photonic crystals on flexible PDMS, diffractive optical elements on curved lenses, spiral phase plates on CMOS chips, multilayer achromatic metalens on optical fiber facet, as well as 3D glass photonic crystals and optical topological resonators on anti-stiction quartz.

physics.optics↗
arXiv2026

Integrated Sensing, Communication, and Computing in Multi-Tier Systems: Joint Hybrid Beamforming Design and Computation Resource Allocation

This paper proposes a novel integrated sensing, communication, and computing (ISCC) framework over a cloud-edge-device collaborative architecture, where passive sensing is enabled by reusing uplink offloading signals to extract sensing information directly at the edge without incurring additional transmission overhead. Nevertheless, such signal reuse introduces an inherent tradeoff between communication efficiency and sensing coverage. To address this challenge, we adopt a hybrid beamforming architecture under practical hardware constraints. In addition, the integration of sensing tasks creates significant resource contention at the mobile edge computing (MEC) server, where latency-sensitive device tasks and computation-intensive sensing inference tasks compete for limited processing capacity. To alleviate this computation burden, we introduce a split inference mechanism that strategically partitions intelligent sensing tasks between the edge and the cloud. Building upon this framework, we formulate a joint optimization problem to minimize the average computation latency of all device tasks subject to strict sensing performance constraints. To tackle the high non-convexity of the formulated problem, we develop an efficient alternating optimization algorithm. In particular, we design a two-layer framework to jointly determine the optimal DNN splitting point and computation resource allocation and employ a weighted minimum mean square error (WMMSE)-based approach with manifold optimization for hybrid beamforming design. Numerical results demonstrate that the proposed framework achieves a superior tradeoff between sensing accuracy and computation latency compared to the benchmark schemes.

eess.SP↗
arXiv2026

Magnetically induced Circular Photogalvanic Effect in Symmetric Two-dimensional Materials

Photocurrents that depend on the helicity of the incident light can be generated in both bulk and low-dimensional materials lacking inversion symmetry, known as the circular photogalvanic effect (CPGE). We propose that by employing a magnetic effect, the limitation on the inversion symmetry broken materials can be overcome, such that helicity-dependent photocurrent can be generated in a symmetric material, i.e., a magneto-circular photogalvanic effect (MCPGE). As a proof of principle, we elucidate the mechanism of such an MCPGE through an effective Hamiltonian of a monolayer SbH on a magnetic substrate with an adjustable magnetization. Moreover, the associated response in optical absorption is analyzed, both single-particle and excitonic, through a Bethe-Salpeter equation to describe the Coulomb interaction in excitons. Our result broadens the mechanism of CPGE and opens new opportunities for optoelectronic devices.

cond-mat.mes-hall↗
arXiv2026

SCOUT: Self-Checking and Recovery-Aware Tool-Thought Agents for Ultra-Long Egocentric Video Reasoning

Ultra-long egocentric video understanding requires reasoning over temporally sparse evidence distributed across hours or days, challenging current multimodal models with limited context and the grounding of key video segments. While Chain-of-Tool-Thought (CoTT) agent systems enable iterative retrieval and inspection, they suffer from error propagation due to rigid zoom-in strategies that lack recovery mechanisms. In this work, we address these challenges through SCOUT (Self-Checking Chain-Of-Tool-thought), a recovery-aware agentic framework introducing an adaptive policy that evaluates intermediate tool observations and dynamically trades off exploitation (zoom-in) and exploration (region switching), enabling robust multi-hop reasoning over extremely long horizons. However, training such multi-turn tool-using agents remains challenging, as existing RL methods rely on sparse outcome-level rewards and lack supervision over extended decision trajectories, resulting in suboptimal credit assignment for long-horizon reasoning. To address this, we develop UPS-GRPO, an uncertainty-prioritized policy optimization method that concentrates exploration on high-uncertainty post-tool states while preserving sample efficiency. We further introduce a turn-level advantage decomposition that integrates outcome rewards with tool-grounded temporal alignment rewards for improved credit assignment. Experiments show that SCOUT achieves state-of-the-art results on ultra-long egocentric benchmarks, while remaining competitive on shorter-horizon long-video settings.

cs.AI↗
arXiv2026

ConceptTree: Bringing Semantic Transparency to Black-Box Decision Making for Robotic Manipulation

Establishing interpretable decision-making processes in long-horizon robotic manipulation is critical for enabling reliable human oversight and intervention. However, existing approaches to robotic manipulation largely treat skill selection as opaque mappings from observations to actions, offering limited transparency into how decisions are formed. In this work, we propose ConceptTree, a framework that reframes high-level manipulation skill selection as reasoning over human-interpretable concepts, representing high-level policies as a sequence of concept-level predicates over visual observations. Rather than relying on implicit latent representations, our method learns a normalized concept space grounded in visual inputs, over which a decision tree is trained to predict high-level skills. This formulation yields a transparent decision process that is both traceable and intervenable, enabling direct inspection and modification of policy behavior. We evaluate our approach on a set of real-world robotic manipulation tasks with increasing complexity. Experimental results show that ConceptTree consistently outperforms existing concept-based baselines, particularly in complex, long-horizon scenarios. Furthermore, we provide qualitative case studies showing that our model supports fine-grained intervention by modifying individual concepts, enabling targeted correction of decision errors without retraining.

cs.RO↗
arXiv2026

Point Tracking in Surgery--The 2025 Surgical Tattoos in Infrared Challenge (STIRC2025)

Point tracking in surgery is crucial to enable applications in downstream tasks such as segmentation, 3D reconstruction, virtual tissue landmarking, autonomous probe-based scanning, and subtask autonomy. This paper introduces the 2025 iteration of a point tracking challenge to address this, wherein participants submit their algorithms for quantification. Their algorithms are evaluated using a dataset named surgical tattoos in infrared (STIR), with the challenge named the STIR Challenge 2025 (STIRC2025). The STIR Challenge 2025 comprises two quantitative components: accuracy and efficiency. The accuracy component tests the accuracy of algorithms on in vivo and ex vivo sequences. The efficiency component tests algorithm inference latency. The challenge was conducted as a part of MICCAI EndoVis 2025, and seven teams participated in this challenge. In this paper we summarize the challenge results and participant methods. The challenge dataset is available at: https://zenodo.org/records/20191078, and the code for baseline models and metrics calculation is available here: https://github.com/athaddius/STIRMetrics

cs.CV↗
arXiv2026

WildTrace: Benchmarking Natural Evidence Trails in Long-Context Reasoning

Answering complex questions over long documents frequently requires integrating evidence that the source itself disperses naturally across distant passages. In an incident report, the operating condition, design flaw, and missed safety check that jointly explain a disaster may appear dozens of sections apart; in a novel, a character's true motive may surface only through scenes far removed from the moment it becomes relevant. This source-internal evidence integration is central to real-world long-document analysis, yet existing benchmarks largely sidestep it. Needle probes, planted facts, and reverse-engineered multi-hop chains embed evidence that may differ from the host text in distribution, placement, or register, making it unclear whether strong performance reflects genuine source reasoning or distributional artifacts. We introduce WILDTRACE, a benchmark of 481 tasks over 214 naturally occurring long-form sources such as technical incident reports and lesser-known literary narratives, where all evidence trails arise from the document's own causal, temporal, and narrative logic. Drawing on Pearl's causal hierarchy and prior multi-hop reasoning typologies, we define seven source-internal evidence geometries that characterize the distinct relational demands of analytical reading in long documents. A source-first construction pipeline mines candidate trails from document structure before writing questions; each item then undergoes multi-stage validation covering clue necessity, answer groundedness, rubric fidelity, contamination resistance and answerability. As models are increasingly entrusted with real-world high-stakes analytical tasks, this gap between accessing information and reasoning over naturally dispersed evidence emerges as a defining challenge for the next stage of long-context research.

cs.CL↗
arXiv2026

Efficient photo-ionizing elimination of detrimental electric fields for Rydberg atoms

Rydberg atoms are highly sensitive to external electric fields due to their exaggerated electronic properties. This unique feature lays the foundation for many of their applications in quantum science. However, an uncontrolled stray electric field can be detrimental, severely degrading their quantum control. In this work, we demonstrate a universal scheme that relies on the efficient creation of an in-vacuum plasma source by photo-ionizing laser-cooled atoms to eliminate detrimental electric fields in a Rydberg-atom tweezer array platform, requiring only readily available resources. With this method, we began with a Stark-ionized Rydberg continuum spectrum caused by a large, unknown stray electric field and ultimately recovered stable, coherent excitation of an individual Rydberg state after fully eliminating the field. Our method is directly applicable to existing Rydberg-atom platforms and can also be useful in other experiments sensitive to stray electric fields.

physics.atom-ph↗
arXiv2026

Imputation Meets Clustering: Exploiting Latent Subgroup Structure for Missing Data Recovery

Missing data is prevalent in practical applications, making effective imputation an essential preprocessing step for downstream analysis. Real-world datasets often exhibit complex latent structures composed of multiple subgroups with distinct distributions. However, existing methods often overlook such population heterogeneity. Without explicit structural guidance, these methods tend to produce generic estimates that blur subgroup boundaries and lack instance-level fidelity. While incorporating subgroup information offers a remedy, it faces a circular dependency: reliable subgroup identification requires complete data, while data completion is the imputation objective itself. To resolve this, we propose CAGI (Cluster-Aware Generative Imputation), a framework that reformulates clustering and imputation as a mutually reinforcing co-optimization process. CAGI employs a ``Partition-Guide-Restore'' strategy where dynamic cluster assignments act as local priors to condition a Generative Adversarial Network. An iterative feedback loop is established to progressively refine both cluster structures and imputed values toward faithful subgroup distributions. To ensure distributional stability, CAGI further employs a multi-level optimization objective combining instance-level reconstruction with distribution-level regularization. Extensive experiments on 14 benchmark datasets with 15 representative baselines demonstrate the superiority of CAGI. The source code is available at: https://github.com/supercocachii/CAGI

cs.LG↗
arXiv2026

Why Machines Misread Pedagogical Quality: Human-Machine Alignment in LLM-Based Pretest Question Evaluation

Designing effective pretest questions is challenging at scale: high-quality questions require careful calibration of openness, cognitive depth, and alignment with learning objectives, yet generating and evaluating them manually is time-consuming. We present an AI-assisted workflow for pretest question development that combines automated generation, rubric-based evaluation, and iterative selection. Because the workflow relies on machine evaluation to filter questions at scale, we investigate the alignment between human and machine judgments across a 2x2 design varying rubric operationalization and evaluation mode. Our findings show that human-machine disagreements are systematic rather than random, that rubric revision has a larger effect on alignment than rationale-first evaluation, and that the two interventions are complementary. These findings highlight that scalable AI-assisted pretesting depends not only on generation capability but on how pedagogical quality is operationalized for machine interpretation.

cs.HC↗
arXiv2026

Decoding Naturalistic Emotion Dynamics from the Brain: An LLM-Enhanced Regression Framework

Decoding emotional states from neural signals has been typically framed as a discrete, single-label classification task based on emotionally stable stimuli, a formulation that oversimplifies the continuous, fluid, and co-occurring nature of human affect. This study reconceptualizes emotion decoding by adopting a multi-target regression framework to track multiple overlapping emotional dimensions as continuous trajectories over time. Leveraging the robust generalization capabilities of Large Language Models (LLMs), we extracted fine-grained, continuous sentiment profiles from a naturalistic auditory narrative, Alice in Wonderland, to serve as scalable proxies for subjective affect from human fMRI dataset. Departing from standard classification paradigms or mass-univariate subtractive contrasts that filter out network dynamics, we leverage regularized and kernel-based machine learning algorithms as continuous estimators to track the magnitude of macroscale neural state variations. We demonstrate that models trained on temporal snapshots of Dynamic Functional Connectivity (DFC) significantly outperform static region-of-interest (ROI) amplitude representations, effectively capturing continuous emotional trajectories under rapidly fluctuating narrative input. Furthermore, by implementing graph-theoretical Explainable AI (XAI) techniques, we deconstruct the underlying predictive features to reveal highly interpretable, emotion-specific topological configurations. Collectively, these results highlight the utility of LLM-automated annotation in affective neuroscience and provide compelling empirical evidence for psychological constructionist frameworks, demonstrating that dynamic, distributed network interactions offer superior explanatory power over strictly locationist accounts of emotion.

cs.LG↗