arXiv subjects

Bowen Sun

Publications and source records attributed to Bowen Sun.

At least 19 recordsLinked to original sources

arXiv2026

Reference-Beam-Assisted Perturbation-Resistant Imaging Through A Single Multimode Fiber

Multimode fiber (MMF) imaging is highly sensitive to mechanical perturbations, which alter modal coupling and disrupt the speckle-to-target mapping. Here, we propose a reference-beam-assisted method for image reconstruction across multiple fiber configurations. An independent reference beam propagates through the same MMF as the imaging signal without target modulation, providing target-independent transmission-state information through its speckle patterns without requiring a separate configuration-sensing structure. A neural network jointly processes the signal and reference speckles to reconstruct target images. Experiments with MNIST and Fashion-MNIST across 10 fiber configurations demonstrate improved reconstruction quality, faster training convergence, and more accurate configuration identification. Further evaluation across 40 configurations confirms the reconstruction benefit of reference-beam assistance over a larger set of perturbed fiber states. These results provide a foundation for developing perturbation-adaptive MMF imaging systems for flexible endoscopy and minimally invasive optical diagnostics.

physics.optics↗

arXiv2026

PRISM: PRior-guided Imagination Sampling in world Models

Latent world models learn representations that enable them to predict how actions change the environment - a capability we call physical intuition. We find that these same representations also support action intuition: a lightweight head can learn a goal-conditioned action prior directly from the frozen encoder's features. We introduce PRISM to investigate how action intuition can be extracted from frozen world-model representations and used to guide goal-conditioned planning. The prior's mean and predicted uncertainty guide the search for useful actions, while the world model evaluates their consequences against the goal. Experiments on four simulated control tasks and two real-robot platforms show that a lightweight action head improves planning performance, and that the same fusion interface accepts a diffusion prior over the same frozen features without modification. These results demonstrate the potential of reusing world-model representations to both predict action consequences and guide action selection.

cs.RO↗

arXiv2026

Resource-based reliability of quantum-coherent devices

Quantum devices can lose the resource needed for a specified function without entering a fixed failure subspace. We propose a resource-based reliability framework for this situation, quantifying the remaining task-relevant resource relative to a stated ideal capacity. The resource measure is selected through its operational connection to the task and the allowed operations. We develop this approach for a trusted quantum random-number generator and a calibrated two-port interferometer. Relative-entropy coherence determines the optimal asymptotic yield of standard coherent qubits under incoherent operations for the former, while logarithmic $l_1$ coherence quantifies the fringe resource of the latter and determines its calibrated correct-port probability. An exactly solvable multi-spin spin-boson model illustrates how environmental records affect these resources at the same isolated-spin dephasing strength. Collective noise preserves coherence within magnetization sectors for product inputs, but enhances the decay of Greenberger-Horne-Zeilinger branch coherence. Finite-record simulations connect the resource definitions to measurement counts and show that increasing the sample size reduces statistical fluctuations without eliminating bias from an incorrect many-spin noise model. The resulting framework provides a task-specific route to reliability assessment when a fixed failure projector is insufficient, while keeping retained resource capacity distinct from a classical survival probability.

quant-ph↗

arXiv2026

Nudge Before You Push: Physics-Aware Navigation via Tactile Probing

Visually identical containers can conceal loads that require different handling decisions. We present TANav, which uses a brief nudge to measure push resistance for navigation under a site-defined handling boundary. TacPhys reads the force sequence, with optional RGB-D and kinematics, into a mass estimate for push authorization. A repeated-patrol planner weighs probe and route costs, requests a second contact when useful, and reuses observations across visits. In simulation, TacPhys approaches a resistance-only Bayes reference and reduces missed pushes from 28.7% to 5.5% relative to peak-force thresholding at comparable low-risk operating points. In repeated-patrol simulation, TANav recovers 90% of the oracle's path saving, more than halves human interventions relative to always-detour, and reduces boundary violations from 4.3% to 2.9% relative to RGB-D-only probing. On a quadruped manipulator with a Hall-array fingertip, offline zero-shot MAE is 0.28-1.07 kg on containers up to 2.82 kg. Force-rise calibration at 3 kg gives 93.5% pooled offline accuracy (86.4% on non-cube episodes); a separate raw-peak rule gives 15 of 20 correct online decisions on unseen boxes.

cs.RO↗

arXiv2026

Resource-Constrained Semantic-Aware Remote Estimation with Overlapping Sensor Coverage

We study semantic-aware remote estimation of multiple finite-state Markov sources observed by K sensors with overlapping coverage. The sensors share a time-division multiple-access uplink and differ in transmission reliability, delivery delay, and transmission budget. In each slot, the scheduler jointly selects a source and one of its monitoring sensors, or remains idle, to minimize the long-run average cost of actuation error subject to global and per-sensor transmission-frequency constraints. We formulate this problem as a finite average-cost constrained Markov decision process. We show that the transmission resource functions have rank at most K, although the global constraint may still restrict the feasible region. Consequently, the Lagrangian depends only on K effective transmission costs, and an optimal constrained solution can be represented using at most K+1 deterministic policy-recurrent-class components. We further characterize the piecewise-affine concave Lagrangian value function and derive projected dual subgradient ascent over an explicitly bounded multiplier set. Numerical results illustrate the value-function structure, the need for policy randomization in a representative instance, and the interaction between global and per-sensor transmission budgets.

cs.IT↗

arXiv2026

Decomposition Attacks Across Unlinkable Identities: Limits of Stateful Defenses for LLM Services

Most large language model services use stateless defenses, which judge only the current request, to refuse harmful tasks. Decomposition attacks exploit this limitation by splitting a harmful task into individually permissible requests and combining their answers. Defending against them therefore requires a stateful monitor that considers requests together. If it can group all requests for one attacker task, it can stop the attack. However, attackers can use unlinkable identities and combine answers elsewhere, leaving no reliable grouping signal. We ask whether decomposition attacks can still be stopped under this setting. For a fixed attack strategy without retries, we prove that the achievable security and utility tradeoff depends entirely on how benign requests for the same capabilities are grouped. Persistent, recognizable groups permit a useful defense; fresh, indistinguishable groups do not. When attackers can retry and learn from Allow/Block decisions, this useful operating point disappears: the feedback reveals what passes but not whether a block was correct. Experiments on 91 executable tasks and 11,393 capability-matched benign requests support these results. Under a 1% denial cap for these requests and a 0.5% cap for unrelated background traffic, all ten tested policies, including one privileged policy with an exact request-to-operation map, either fail to stop attacks or exceed the budget. On defense-unseen task families, attack success is at least 99% after one attempt and 100% after two. Effective defenses therefore require additional evidence or mechanisms tied to grouping, such as reliable identity linkage, costs for fresh identities, or control over answer use.

cs.CR↗

arXiv2026

KeyPooling: Measuring Where LLM API Relay Paths Collapse Prompt Cache Isolation

Large language model (LLM) API relays authenticate customers separately but often forward requests through shared provider credentials. Providers scope prompt caches to upstream principals and namespaces, so relay customers mapped to one cache identity can observe each other's cache state. Prior work showed cache sharing at selected endpoints but did not identify which credential, pool, adapter, or nested hop controls the finalidentity. We present KeyPooling, a measurement method that traces customer identity through cache lookup and write, verifies runtime transformations, and tests one predicted identity component at a time. Across five open-source gateways connected to OpenAI and Anthropic, none bound customers to upstream credentials by default; under a shared credential, all five exposed cross-customer cache reads for both providers. Principal and namespace splits, pool associations, and adapter and nested-relay contrasts localized the controlling transformations. In an outcome-independent weekly OpenRouter frame, tests covered 80.5% of eligible token volume and found cross-account reads for 12 of 28 labels carrying 33.7% of volume. On one production route, a controlled procedure recovered eight consecutive target positions without target access. Broader tests identify cache granularity, routing, rate limits, attribution, and budget as conditions for token-by-token recovery, not security controls. We derive a defense contract: every customer must enter a provider-enforced domain, or a namespace derived from authenticated identity must survive every final cache lookup and write. Placing this split after reusable public prefixes preserved most modeled reuse at a 1.7-2.5% cost increase.

cs.CR↗

arXiv2026

An improved bond-associated peridynamic model and its adaptive coupling with CCM for fracture analysis

This paper reformulates the correction factor in the force-state of the bond-associated peridynamic (BAPD) model. The reformulation is established from the strain energy density equivalence between the BAPD model and the classical continuum mechanics (CCM) model at a material point. With the FEM solution taken as the reference, the proposed correction factor improves the accuracy of the BAPD solution in this study. Furthermore, a BAPD-CCM coupled model is developed based on the above energy density equivalence, and a ``Morphing" function is introduced to achieve a smooth transition between the two models. For time integration, explicit schemes are adopted for both quasi-static and dynamic problems. In the spatial discretization, the CCM model is discretized by elements, whereas the BAPD model is discretized by particles. The solution accuracy of the coupled model is validated by comparison with the FEM solution and by evaluating the $L^{2}$ norm, the $H^{1}_{\mathrm{semi}}$, and the energy norm of the displacement error. Two- and three-dimensional numerical examples show that the proposed model has higher computational efficiency. For example, in the Mode I crack propagation problem, its computational cost is reduced by more than 72\% compared with that of the pure BAPD model, and the predicted crack patterns agree with experimental results.

cs.CE↗

arXiv2026

Automated High-Precision Extraction and Forensic Verification of Data-Bearing Vector Figures

The quantitative record of science and engineering is increasingly carried by figures rather than text or tables, and a reader who needs the underlying numbers must usually re-digitize them by hand: slowly, imprecisely, and with no way to prove the result is faithful. Yet when a figure is stored as vector graphics, its data are not approximated by the picture but encoded in it: the renderer writes each marker and vertex at a printed precision that, for the dominant scientific toolchain, exceeds the data's own. We turn this into three contributions, one per shortcoming of hand digitization. First, a precision theory bounding how accurately data can be recovered for a given renderer and export format: bit-exact float32 for matplotlib markers, and a calibration-limited three to four significant figures end to end. Second, an automatic extractor that decodes a figure in one pass with no human in the loop, in place of the slow point-by-point tracing a digitizer demands. Third, a verification theory: recovery is injective except on a characterized, vanishingly small interval near zero; accidental agreement between unrelated data is astronomically unlikely; and a re-rendering certificate binds the recovered values to the markers, lines, and ticks the figure draws, not its text, making a result non-repudiable. With no ground truth used during recovery, decoded figures match external archives (Planck 2018 to 10^-9; the Keeling CO2 record to 5*10^-4, and one decoded figure independently reproduces a correction to the Chinchilla scaling-law confidence interval. We map the achievable precision across common renderers and their PDF, SVG, and EPS formats. What we deliver is certified data; the scientific significance of any particular dataset lies outside this paper's scope, and recovered values are candidates for human review, never accusations.

cs.CR↗

arXiv2026

Comb-enabled spectral-domain image transport through perturbation-prone multimode fibers

Multimode fibers (MMFs) offer a compact platform for imaging, sensing, and information transport, but their practical deployment is hindered by sensitivity to fiber perturbations, which alter modal coupling and invalidate conventional speckle-based calibrations. Here, we demonstrate perturbation-resilient image transport through MMFs by combining image-to-spectrum encoding with dual-comb spectroscopy. Two-dimensional images are converted into comb-line-resolved spectral signatures before fiber transmission, allowing spatial information to be carried in the spectral domain rather than in the output speckle field. After propagation, dual-comb heterodyne detection maps the encoded spectrum into the radio-frequency domain, enabling massively parallel spectral readout with a single photodetector. Neural-network-assisted compressive reconstruction further enables high-fidelity imaging from sparse, noisy, and spectrally aliased measurements. Our approach achieves Pearson correlation coefficients exceeding 0.9 under strong fiber perturbations and supports frame rates up to 2.5 MHz, allowing the observation of transient switching dynamics in a digital micromirror device. These results establish a powerful tool for robust, real-time image transport through flexible MMFs, with potential applications in remote sensing and fiber-based optical instrumentation.

physics.optics↗

arXiv2026

TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning

Decompositional jailbreaks pose a critical threat to large language models (LLMs) by allowing adversaries to fragment a malicious objective into a sequence of individually benign queries that collectively reconstruct prohibited content. In real-world deployments, LLMs face a continuous, untraceable stream of fully anonymized and arbitrarily interleaved requests, infiltrated by covertly distributed adversarial queries. Under this rigorous threat model, state-of-the-art defensive strategies exhibit fundamental limitations. In the absence of trustworthy user metadata, they are incapable of tracking global historical contexts, while their deployment of generative models for real-time monitoring introduces computationally prohibitive overhead. To address this, we present TwinGate, a stateful dual-encoder defense framework. TwinGate employs Asymmetric Contrastive Learning (ACL) to cluster semantically disparate but intent-matched malicious fragments in a shared latent space, while a parallel frozen encoder suppresses false positives arising from benign topical overlap. Each request requires only a single lightweight forward pass, enabling the defense to execute in parallel with the target model's prefill phase at negligible latency overhead. To evaluate our approach and advance future research, we construct a comprehensive dataset of over 3.62 million instructions spanning 8,600 distinct malicious intents. Evaluated on this large-scale corpus under a strictly causal protocol, TwinGate achieves high malicious intent recall at a remarkably low false positive rate while remaining highly robust against adaptive attacks. Furthermore, our proposal substantially outperforms stateful and stateless baselines, delivering superior throughput and reduced latency.

cs.CR↗

arXiv2026

Some Large Language Models Exhibit Consistent Risk Attitudes

As artificial intelligence systems are deployed in open-ended, high-stakes settings, a critical dimension remains unmeasured: how perceived risk is translated into action. We test whether large language models (LLMs) exhibit systematic and consistent risk attitudes under uncertainty. We introduce a cross-domain framework that decouples contextual risk belief from categorical decision, and apply it to six representative LLMs and 100 human participants across spatial navigation, clinical triage, and financial allocation tasks. Using regression models, we extract each agents belief-to-decision mapping and quantify risk sensitivity and risk attitude bias. We find that most tested LLMs exhibit (i) robust intra-task consistency, indicating stable mappings from contextual belief to risk decision within a fixed task domain; (ii) cross-domain rank-order stability, preserving relative risk posture across tasks; and (iii) a convergence toward a restricted risk-attitude distribution relative to the broader human baseline. These results reveal risk attitude as a stable and previously uncharacterized dimension of LLM behavior, establishing a foundation for evaluating and aligning AI systems in open-ended decision-making and motivating further investigation into the origins of these intrinsic behavioral dispositions.

cs.AI↗

arXiv2026

CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning

Understanding camera dynamics is a fundamental pillar of video spatial intelligence. However, existing multimodal models predominantly treat this task as a black-box classification, often confusing physically distinct motions by relying on superficial visual patterns rather than geometric cues. We present \textbf{CamReasoner}, a framework that reformulates camera movement understanding as a structured inference process to bridge the gap between perception and cinematic logic. Our approach centers on the Observation-Thinking-Answer (O-T-A) paradigm, which compels the model to articulate spatio-temporal observations and reason about motion patterns within an explicit reasoning block. To instill this capability, we construct a Large-scale Inference Trajectory Suite comprising 18k SFT reasoning chains and 38k RL feedback samples. To the best of our knowledge, \textbf{we are the first to employ RL for logical alignment in camera movement understanding}, ensuring motion inferences are grounded in structured visual reasoning rather than contextual guesswork. Built upon Qwen2.5-VL-7B, CamReasoner-7B improves binary classification accuracy from 73.8\% to 78.4\% and VQA accuracy from 60.9\% to 74.5\% over its backbone, consistently outperforming both proprietary and open-source baselines across multiple benchmarks.

cs.CV↗

arXiv2026

PAS: A Training-Free Stabilizer for Temporal Encoding in Video LLMs

Video LLMs suffer from temporal inconsistency: small shifts in frame timing can flip attention and suppress relevant frames. We trace this instability to the common extension of Rotary Position Embeddings to video through multimodal RoPE. The induced inverse Fourier time kernel exhibits frame-scale ripples that multiply adjacent frames by different factors, which perturbs attention that should otherwise be governed by the raw query key inner product. We present Phase Aggregated Smoothing (PAS), a simple, training-free mechanism that applies small opposed phase offsets across heads and then aggregates their outputs. PAS preserves the per-head spectrum magnitude, while the aggregation effectively smooths the temporal kernel and reduces phase sensitivity without changing the positional encoding structure. Our analysis shows that the RoPE rotated logit can be approximated as a content dot product scaled by a time kernel; smoothing this kernel yields Lipschitz stability of attention to small temporal shifts; multi phase averaging attenuates high frequency ripples while preserving per-head spectra under Nyquist-valid sampling. Experiments on multiple video understanding benchmarks under matched token budgets show consistent improvements with negligible computational overhead. PAS provides a plug and play upgrade for robust temporal encoding in Video LLMs.

cs.CV↗

arXiv2026

Reliability Dynamics in a Two-Site Dissipative Quantum Spin Chain

As a key index for applications of a device, the device's reliability is its ability to survive (function normally over time) under the influence of some environment. In this paper we present a quantum energy-storing device model with a quantum spin chain, whose environment influence is described by the Lindblad master equation. Here the device survives if the spin system stays in the state with nonzero excitations; otherwise, it fails. Because the Lindblad dynamics enforces one-way energy decay and strict irreversibility of the failure state, we can investigate the reliability of the quantum device directly using classical reliability theory. Focusing on the minimal nontrivial case -- a two-site spin-1/2 chain -- we derive closed-form expressions for the reliability and the hazard rate. The dynamics exhibit an overdamped-underdamped crossover controlled by the competition between coherent exchange and dissipation inhomogeneity. The exact analytical formulas are in excellent agreement with numerical simulations. More importantly, we establish an experimentally accessible protocol for assessing reliability based on first-passage time statistics.

quant-ph↗

arXiv2026

Green or Fast? Learning to Balance Cold Starts and Idle Carbon in Serverless Computing

Serverless computing simplifies cloud deployment but introduces new challenges in managing service latency and carbon emissions. Reducing cold-start latency requires retaining warm function instances, while minimizing carbon emissions favors reclaiming idle resources. This balance is further complicated by time-varying grid carbon intensity and varying workload patterns, under which static keep-alive policies are inefficient. We present LACE-RL, a latency-aware and carbon-efficient management framework that formulates serverless pod retention as a sequential decision problem. LACE-RL uses deep reinforcement learning to dynamically tune keep-alive durations, jointly modeling cold-start probability, function-specific latency costs, and real-time carbon intensity. Using the Huawei Public Cloud Trace, we show that LACE-RL reduces cold starts by 51.69% and idle keep-alive carbon emissions by 77.08% compared to Huawei's static policy, while achieving better latency-carbon trade-offs than state-of-the-art heuristic and single-objective baselines, approaching Oracle performance.

cs.DC↗

arXiv2025

From Chaos to Clarity: Schema-Constrained AI for Auditable Biomedical Evidence Extraction from Full-Text PDFs

Biomedical evidence synthesis relies on accurate extraction of methodological, laboratory, and outcome variables from full-text research articles, yet these variables are embedded in complex scientific PDFs that make manual abstraction time-consuming and difficult to scale. Existing document AI systems remain limited by OCR errors, long-document fragmentation, constrained throughput, and insufficient auditability for high-stakes synthesis. We present a schema-constrained AI extraction system that transforms full-text biomedical PDFs into structured, analysis-ready records by explicitly restricting model inference through typed schemas, controlled vocabularies, and evidence-gated decisions. Documents are ingested using resume-aware hashing, partitioned into caption-aware page-level chunks, and processed asynchronously under explicit concurrency controls. Chunk-level outputs are deterministically merged into study-level records using conflict-aware consolidation, set-based aggregation, and sentence-level provenance to support traceability and post-hoc audit. Evaluated on a corpus of studies on direct oral anticoagulant level measurement, the pipeline processed all documents without manual intervention, maintained stable throughput under service constraints, and exhibited strong internal consistency across document chunks. Iterative schema refinement substantially improved extraction fidelity for synthesis-critical variables, including assay classification, outcome definitions, follow-up duration, and timing of measurement. These results demonstrate that schema-constrained, provenance-aware extraction enables scalable and auditable transformation of heterogeneous scientific PDFs into structured evidence, aligning modern document AI with the transparency and reliability requirements of biomedical evidence synthesis.

cs.CL↗

arXiv2025

Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding

Discrete diffusion language models have shown strong potential for text generation, yet standard supervised fine-tuning (SFT) misaligns with their semi-autoregressive inference: training randomly masks tokens across the entire response, while inference generates fixed-size blocks sequentially. This mismatch introduces noisy prefixes and leaky suffixes, biasing gradients away from the desired blockwise likelihood. We propose Blockwise SFT, which partitions responses into fixed-size blocks, selects one active block per step for stochastic masking, freezes all preceding tokens, and fully hides future ones. Loss is computed only over the active block, directly mirroring the blockwise decoding process. Experiments on GSM8K, MATH, and MetaMathQA show consistent gains over classical SFT under equal compute or token budgets. Block size consistency studies and ablations confirm that improvements stem from faithful training-inference alignment rather than incidental masking effects. Our results highlight the importance of matching supervision granularity to the decoding procedure in diffusion-based language models.

cs.CL↗