arXiv subjects

Zhen Zhang

Publications and source records attributed to Zhen Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2026

Data-Driven Tube-Based Zonotopic Predictive Control With Nonconvex Layered Terminal Sets

This paper presents a data-driven tube-based zonotopic predictive control (DTZPC) framework with nonconvex layered terminal sets. Existing DTZPC schemes with closed-loop guarantees typically rely on a single ellipsoidal terminal set, which can be conservative and thereby limit feasibility. We propose a layered terminal-set design that decouples stability certification, feasibility enlargement, and motion-region screening into three components with distinct roles. First, an offline-designed feedback gain together with a contractive constrained zonotope provides a terminal ingredient for stability certification, while avoiding probabilistic feedback synthesis in high-dimensional DTZPC. Second, we derive a data-driven characterization of the inverse admissible closed-loop model set, avoiding the conservatism of interval-matrix relaxation and inversion. Combined with exact set multiplication, this yields inner and outer approximations of the maximal robust positively invariant (MRPI) set under fixed closed-loop dynamics. The inner approximation serves as a nonconvex terminal set to enlarge feasibility, whereas the outer approximation provides certified motion-region descriptions for fast screening and monitoring. Numerical examples demonstrate tighter inverse-set enclosures and improved feasibility over existing convex-terminal DTZPC schemes.

math.OC↗

arXiv2026

Transformer-Enhanced Data-Driven Output Reachability with Conformal Coverage Guarantees

This paper considers output reachability analysis for linear time-invariant systems with unknown state-space matrices and unknown observation map, given only noisy input-output measurements. The Cayley--Hamilton theorem is applied to eliminate the latent state algebraically, producing an autoregressive input-output model whose parameter uncertainty is enclosed in a matrix zonotope. Set-valued propagation of this model yields output reachable sets with deterministic containment guarantees under a bounded aggregated residual assumption. The conservatism inherent in the lifted matrix-zonotope product is then mitigated by a decoder-only Transformer trained on labels obtained through directional contraction of the formal envelope via an exterior non-reachability certificate. Split conformal prediction restores distribution-free coverage at both per-step and trajectory levels without access to the true reachable-set hull. The framework is validated on a five-dimensional system with multiple unknown observation matrices.

eess.SY↗

arXiv2026

GECAM discovery of a peculiar magnetar X-ray burst (MXB 221120) from SGR J1935+2154 associated with a fast radio burst

Fast radio bursts (FRBs) are enigmatic cosmic transients of millisecond duration observed in the radio band. The identification of FRB-associated magnetar X-ray bursts (MXBs) from galactic magnetar SGR J1935+2154 suggests that at least a fraction of FRBs can be produced from magnetar activity. However, the sample size of FRB-associated MXBs is still very small. Here we report a bright and peculiar FRB-associated MXB from SGR J1935+2154 detected by GECAM on November 20, 2022, dubbed MXB 221120. We find that both temporal and spectral properties of MXB 221120 exhibit distinctive features. Its light curve could be generally described by a single FRED function with superposition of several narrow pulses. Interestingly, we identify a possible QPO feature with center frequency of ~18 Hz in this MXB. The time-integrated spectrum is best fitted by a blackbody model with temperature (kT ) of 18.6 keV, rendering it the first thermal spectrum FRB-associated MXB from SGR J1935+2154. Compared to other MXBs with single emission episode, MXB 221120 has longer duration and higher blackbody temperature, making it an outlier in the burst sample. These results indicate that MXB 221120 may be produced by a special mechanism with extreme physical conditions.

astro-ph.HE↗

arXiv2026

When Users Change Their Mind: Evaluating Interruptible Agents in Long-Horizon Web Navigation

As LLM agents transition from short, static problem solving to executing complex, long-horizon tasks in dynamic environments, the ability to handle user interruptions, such as adding requirement or revising goals, during mid-task execution is becoming a core requirement for realistic deployment. However, existing benchmarks largely assume uninterrupted agent behavior or study interruptions only in short, unconstrained language tasks. In this paper, we present the first systematic study of interruptible agents in long-horizon, environmentally grounded web navigation tasks, where actions induce persistent state changes. We formalize three realistic interruption types, including addition, revision, and retraction, and introduce InterruptBench, a benchmark derived from WebArena-Lite that synthesizes high-quality interruption scenarios under strict semantic constraints. Using a unified interruption simulation framework, we evaluate six strong LLM backbones across single- and multi-turn interruption settings, analyzing both their effectiveness in adapting to updated intents and their efficiency in recovering from mid-task changes. Our results show that handling user interruptions effectively and efficiently during long-horizon agentic tasks remains challenging for powerful large-scale LLMs. Code and dataset are available at https://github.com/HenryPengZou/InterruptBench.

cs.CL↗

arXiv2026

Certified Set Convergence for Piecewise Affine Systems via Neural Lyapunov Functions

Safety-critical control of piecewise affine (PWA) systems under bounded additive disturbances requires guarantees not for individual states but for entire state sets simultaneously: a single control action must steer every state in the set toward a target, even as sets crossing mode boundaries split and evolve under distinct affine dynamics. Certifying such set convergence via neural Lyapunov functions couples the Lipschitz constants of the value function and the policy, yet certified bounds for expressive networks exceed true values by orders of magnitude, creating a certification barrier. We resolve this through a three-stage pipeline that decouples verification from the policy. A value function from Hamilton-Jacobi backward reachability, trained via reinforcement learning, is the Lyapunov candidate. A permutation-invariant Deep Sets controller, distilled via regret minimization, produces a common action. Verification propagates zonotopes through the value network, yielding verified Lyapunov upper bounds over entire sets without bounding the policy Lipschitz constant. On four benchmarks up to dimension six, including systems with per-mode operator norms exceeding unity, the framework certifies set convergence with positive margin on every system. A spectrally constrained local certificate completes the terminal guarantee, and the set-actor is the only tested method to achieve full strict set containment, at constant-time online cost.

eess.SY↗

arXiv2026

Metallic d-wave altermagnetism in WFeB: a platform for electrically switchable perpendicular spin-splitter response

We report the synthesis and magnetic characterization of WFeB and identify it as a metallic d-wave altermagnet representative of a broader TiNiSi-type family. Neutron diffraction, Mössbauer spectroscopy, and magnetometry establish a collinear altermagnetic ordering confirmed by first-principles calculations. The electronic structure shows a nonrelativistic spin splitting of approximately 100 meV, but it also supports a strong spin-splitter transport response. This demonstrates that efficient spin-current generation can occur even with such modest band splitting. Symmetry analysis shows that selected film orientations permit deterministic switching of the Néel vector by current-induced staggered torques, enabling electrical control of a perpendicular spin-splitter response. These results establish WFeB and related TiNiSi-type antiferromagnets as a platform for electrically switchable charge-to-spin conversion driven by altermagnetic symmetry.

cond-mat.mtrl-sci↗

arXiv2026

MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome

Recent progress in deep research systems has been impressive, but evaluation still lags behind real user needs. Existing benchmarks predominantly assess final reports using fixed rubrics, failing to evaluate the underlying research process. Most also offer limited multimodal coverage, rely on synthetic tasks that do not reflect real-world query complexity, and cannot be refreshed as knowledge evolves. To address these gaps, we introduce MiroEval, a benchmark and evaluation framework for deep research systems. The benchmark comprises 100 tasks (70 text-only, 30 multimodal), all grounded in real user needs and constructed via a dual-path pipeline that supports periodic updates, enabling a live and evolving setting. The proposed evaluation suite assesses deep research systems along three complementary dimensions: adaptive synthesis quality evaluation with task-specific rubrics, agentic factuality verification via active retrieval and reasoning over both web sources and multimodal attachments, and process-centric evaluation audits how the system searches, reasons, and refines throughout its investigation. Evaluation across 13 systems yields three principal findings: the three evaluation dimensions capture complementary aspects of system capability, with each revealing distinct strengths and weaknesses across systems; process quality serves as a reliable predictor of overall outcome while revealing weaknesses invisible to output-level metrics; and multimodal tasks pose substantially greater challenges, with most systems declining by 3 to 10 points. The MiroThinker series achieves the most balanced performance, with MiroThinker-H1 ranking the highest overall in both settings. Human verification and robustness results confirm the reliability of the benchmark and evaluation framework. MiroEval provides a holistic diagnostic tool for the next generation of deep research agents.

cs.AI↗

arXiv2026

Comprehensive Measurement of Spectral Evolution in a GRB Flare: High Time-Resolution Insights into the "Double-Tracking" Phenomenon

The spectral evolution characteristics of the prompt emission in gamma-ray bursts (GRBs) have been extensively studied, but detailed investigations of spectral evolution in a GRB flare remain lacking. In this work, we present the first analysis of spectral parameter evolution in a GRB flare through high time-resolved spectral fitting of the Brightest Flare in GRB 221009A. We find that the $α$-Flux, $E_p$-Flux, and $E_p$-$α$ relationships during both the overall phase and the rise phase of flare can be well described by simple power-law model, showing positive correlations. Therefore, we conclude that Brightest Flare exhibits "Double-tracking" behavior. Since values of $α$ do not exceed the synchrotron "death line" (-2/3), we explain this phenomenon using a magnetic dissipation synchrotron radiation model. In the decay phase of flare, the $E_p$-Flux and $E_p$-$α$ correlations become notably flatter, with their power-law indices decreasing significantly compared to those in the rise phase. This may be due to the fact that the next flare begins to erupt before the Brightest Flare has completely ended, resulting in the combined effects of both two flares. Our study of spectral parameter relations of the Brightest Flare provides new insights into the radiation mechanisms of both GRB prompt emission and flares.

astro-ph.HE↗

arXiv2026

AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents

As large language models (LLMs) evolve into autonomous agents for long-horizon information-seeking, managing finite context capacity has become a critical bottleneck. Existing context management methods typically commit to a single fixed strategy throughout the entire trajectory. Such static designs may work well in some states, but they cannot adapt as the usefulness and reliability of the accumulated context evolve during long-horizon search. To formalize this challenge, we introduce a probabilistic framework that characterizes long-horizon success through two complementary dimensions: search efficiency and terminal precision. Building on this perspective, we propose AgentSwing, a state-aware adaptive parallel context management routing framework. At each trigger point, AgentSwing expands multiple context-managed branches in parallel and uses lookahead routing to select the most promising continuation. Experiments across diverse benchmarks and agent backbones show that AgentSwing consistently outperforms strong static context management methods, often matching or exceeding their performance with up to $3\times$ fewer interaction turns while also improving the ultimate performance ceiling of long-horizon web agents. Beyond the empirical gains, the proposed probabilistic framework provides a principled lens for analyzing and designing future context management strategies for long-horizon agents.

cs.CL↗

arXiv2026

Advanced Control of Electron Beams: Tailoring X-ray Production with Programmable Laser Shaping

Leveraging the full scientific capabilities of next-generation high-repetition-rate free-electron lasers requires programmable control over electron-beam properties at their source. The photoinjector drive laser defines the electron beam's initial six-dimensional phase-space distribution, yet has historically been limited to Gaussian or static flat-top profiles, with most manipulation occurring downstream. Here we demonstrate software-programmable ultraviolet pulse shaping at the LCLS-II photoinjector as a source-level actuator that complements traditional accelerator controls. Using a coupled architecture combining dispersion-controlled nonlinear frequency conversion with spatial-light-modulator spectral shaping, we generate user-defined temporal structures and observe their imprint on electron bunches through high-resolution time-domain diagnostics. Laser-imposed multi-peaked modulation persists through acceleration, magnetic compression, and undulator transport with shot-to-shot repeatability, producing clearly resolved current structure in the compressed beam. Variance-based reconstruction from transverse deflecting cavity measurements reveals structured X-ray emission profiles exhibiting temporal features consistent with the programmed laser waveform. By providing rapid, software-controlled reconfiguration of electron-beam initial conditions, this source-level control approach establishes a programmable upstream actuator for future adaptive optimization and autonomous facility operation at high-repetition-rate light sources.

physics.acc-ph↗

arXiv2026

Conformalized Data-Driven Reachability Analysis with PAC Guarantees

Data-driven reachability analysis computes over-approximations of reachable sets directly from noisy data. Existing deterministic methods require either known noise bounds or system-specific structural parameters such as Lipschitz constants. We propose Conformalized Data-Driven Reachability (CDDR), a framework that provides Probably Approximately Correct (PAC) coverage guarantees through the Learn Then Test (LTT) calibration procedure, requiring only that calibration and test trajectories be independently and identically distributed. CDDR is developed for three settings: linear time-invariant (LTI) systems with unknown process noise distributions, LTI systems with bounded measurement noise, and general nonlinear systems including non-Lipschitz dynamics. Experiments on a 5-dimensional LTI system under Gaussian and heavy-tailed Student-t noise and on a 2-dimensional non-Lipschitz system with fractional damping demonstrate that CDDR achieves valid coverage where deterministic methods do not provide formal guarantees. Under anisotropic noise, a normalized score function reduces the reachable set volume while preserving the PAC guarantee.

eess.SY↗

arXiv2026

Chem4DLLM: 4D Multimodal LLMs for Chemical Dynamics Understanding

Existing chemical understanding tasks primarily rely on static molecular representations, limiting their ability to model inherently dynamic phenomena such as bond breaking or conformational changes, which are essential for a chemist to understand chemical reactions. To address this gap, we introduce Chemical Dynamics Understanding (ChemDU), a new task that translates 4D molecular trajectories into interpretable natural-language explanations. ChemDU focuses on fundamental dynamic scenarios, including gas-phase and catalytic reactions, and requires models to reason about key events along molecular trajectories, such as bond formation and dissociation, and to generate coherent, mechanistically grounded narratives. To benchmark this capability, we construct Chem4DBench, the first dataset pairing 4D molecular trajectories with expert-authored explanations across these settings. We further propose Chem4DLLM, a unified model that integrates an equivariant graph encoder with a pretrained large language model to explicitly capture molecular geometry and rotational dynamics. We hope that ChemDU, together with Chem4DBench and Chem4DLLM, will stimulate further research in dynamic chemical understanding and multimodal scientific reasoning.

cs.LG↗

arXiv2026

Topological observables and domain wall tension from finite temperature chiral perturbation theory

Within the framework of SU(2) chiral perturbation theory, we derive the general solution of the QCD $θ$-vacuum for an arbitrary vacuum phase, explicitly incorporating isospin-breaking effects from the light quark mass difference, and compute the temperature dependence of the topological susceptibility, higher-order cumulants, and the domain wall tension up to next-to-leading order. We find that the topological susceptibility agrees with lattice data at low temperatures but deviates at higher temperatures as expected from the breakdown of the chiral expansion; moreover, we demonstrate that the normalized fourth-order cumulant and the domain wall tension decrease monotonically with increasing temperature, while the normalized sixth-order cumulant exhibits the opposite behavior. These results extend earlier analyses by showing how isospin breaking reshapes the full hierarchy of topological charge cumulants and the dynamics of $θ$-vacuum domain walls, thereby offering new theoretical input on the $θ$-vacuum properties, which are relevant for axion-related effective theories in hot QCD matter.

hep-ph↗

arXiv2026

DAWA: Dynamic Ambiguity-Wise Adaptation for Real-Time Domain Adaptive Semantic Segmentation

Test-time domain adaption (TTDA) for semantic segmentation aims to adapt a segmentation model trained on a source domain to a target domain for inference on-the-fly, where both efficiency and effectiveness are critical. However, existing TTDA methods either rely on costly frame-wise optimization or assume unrealistic domain shifts, resulting in poor adaptation efficiency and continuous semantic ambiguities. To address these challenges, we propose a real-time framework for TTDA semantic segmentation, called Dynamic Ambiguity-Wise Adaptation (DAWA), which adaptively detects domain shifts and dynamically adjusts the learning strategies to mitigate continuous ambiguities in the test time. Specifically, we introduce the Dynamic Ambiguous Patch Mask (DAP Mask) strategy, which dynamically identifies and masks highly disturbed regions to prevent error accumulation in ambiguous classes. Furthermore, we present the Dynamic Ambiguous Class Mix (DAC Mix) strategy that leverages vision-language models to group semantically similar classes and augment the target domain with a meta-ambiguous class buffer. Extensive experiments on widely used TTDA benchmarks demonstrate that DAWA consistently outperforms state-of-the-art methods, while maintaining real-time inference speeds of approximately 40 FPS.

cs.CV↗

arXiv2026

Eclipsed X-ray Bursts from Magnetar SGR J1935+2154 and the Fireball Measurements

In theory, burst activity of the magnetar can lead to the formation of fireballs trapped by the magnetic field and corotating with the star. However, the smoking-gun observational evidence of the fireball is elusive. We envisage that the fireball emission should occasionally be eclipsed by the magnetar, especially when the burst duration is comparable to the magnetar's spin period. In this work, we first discover a peculiar type of burst whose light curve has a plateau-like feature among the long bursts of the magnetar SGR J1935+2154 detected by GECAM and Fermi/GBM. Then, based on these bursts, we identified four burst candidates with eclipse-like characteristics. By fitting their light curves with the eclipse fireball model, the viewing angle of the magnetar relative to its spin axis is estimated to be $17^\circ \pm 10^\circ$, and the distances from the fireballs to the magnetar are found to be more than 5 times the magnetar's radius, indicating that the fireballs are suspended in the magnetosphere rather than adhering to the magnetar surface. Furthermore, we find that this configuration is well consistent with the implication of the cyclotron resonance scattering feature we found in their spectra. Our results suggest that some intermediate X-ray bursts may originate from magnetic reconnection within the magnetosphere rather than the starquake.

astro-ph.HE↗

arXiv2026

Truth as a Trajectory: What Internal Representations Reveal About Large Language Model Reasoning

Existing explainability methods for Large Language Models (LLMs) typically treat hidden states as static points in activation space, assuming that correct and incorrect inferences can be separated using representations from an individual layer. However, these activations are saturated with polysemantic features, leading to linear probes learning surface-level lexical patterns rather than underlying reasoning structures. We introduce Truth as a Trajectory (TaT), which models the transformer inference as an unfolded trajectory of iterative refinements, shifting analysis from static activations to layer-wise geometric displacement. By analyzing displacement of representations across layers, TaT uncovers geometric invariants that distinguish valid reasoning from spurious behavior. We evaluate TaT across dense and Mixture-of-Experts (MoE) architectures on benchmarks spanning commonsense reasoning, question answering, and toxicity detection. Without access to the activations themselves and using only changes in activations across layers, we show that TaT effectively mitigates reliance on static lexical confounds, outperforming conventional probing, and establishes trajectory analysis as a complementary perspective on LLM explainability.

cs.CL↗

arXiv2026

I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?

Recent empirical evidence shows that LLM representations encode human-interpretable concepts. Nevertheless, the mechanisms by which these representations emerge remain largely unexplored. To shed further light on this, we introduce a novel generative model that generates tokens on the basis of such concepts formulated as latent discrete variables. Under mild conditions, even when the mapping from the latent space to the observed space is non-invertible, we establish rigorous identifiability result: the representations learned by LLMs through next-token prediction can be approximately modeled as the logarithm of the posterior probabilities of these latent discrete concepts given input context, up to an linear transformation. This theoretical finding: 1) provides evidence that LLMs capture essential underlying generative factors, 2) offers a unified and principled perspective for understanding the linear representation hypothesis, and 3) motivates a theoretically grounded approach for evaluating sparse autoencoders. Empirically, we validate our theoretical results through evaluations on both simulation data and the Pythia, Llama, and DeepSeek model families.

cs.LG↗

arXiv2026

Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning

Directed Acyclic Graphs (DAGs) are a standard tool in causal modeling, but their suitability for capturing the complexity of large-scale multimodal data is questionable. In practice, real-world multimodal datasets are often collected from heterogeneous generative processes that do not conform to a single DAG. Instead, they may involve multiple, and even opposing, DAG structures with inverse causal directions. To address this gap, in this work, we first propose a novel latent partial causal model tailored for multimodal data representation learning, featuring two latent coupled variables parts connected by an undirected edge, to represent the transfer of knowledge across modalities. Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by MultiModal Contrastive Learning (MMCL) correspond to the latent coupled variables up to a trivial transformation. This result deepens our understanding of the why MMCL works, highlights its potential for representation disentanglement, and expands the utility of pre-trained models like CLIP. Synthetic experiments confirm the robustness of our findings, even when the assumptions are partially violated. Most importantly, experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets. Together, these contributions push the boundaries of MMCL, both in theory and in practical applications.

cs.LG↗