Search arXiv⌕ Search

arXiv subjects

Pengfei Zhang

Publications and source records attributed to Pengfei Zhang.

At least 19 recordsLinked to original sources
arXiv2026

The Kubo-Thermalization Correspondence

Quantum thermalization describes how interacting quantum systems relax toward thermal equilibrium, a central problem in modern physics. Yet most experimental information on many-body systems comes from short-time transition spectroscopy, typically interpreted within Kubo's linear-response framework. These perspectives-long-time equilibration versus short-time response-seem fundamentally disconnected. Here we establish an exact link between them: the \emph{Kubo-Thermalization correspondence}, which connects long-time thermalized magnetization under weak driving to short-time linear-response spectra for a spin coupled to a thermal bath. The correspondence holds even when the steady state differs substantially from the initial state and when each regime is individually difficult to describe theoretically. We experimentally confirm the correspondence using effective spin-$1/2$ impurities realized with ultracold fermions in two internal states coupled to a Fermi sea. Our results provide a rare exact statement about quantum thermalization and offer a novel route to infer thermalization dynamics from equilibrium response measurements in strongly interacting quantum systems, independent of microscopic details of the system--bath coupling.

cond-mat.quant-gas↗
arXiv2026

Characterizing Fermionic Non-Gaussianity in the Sachdev-Ye-Kitaev Model via Replica Twist Entropy

Understanding the non-equilibrium dynamics of quantum many-body systems plays a central role in modern many-body physics. Intriguing insights emerge from investigating how quantum resources, which are essential for achieving unambiguous quantumness, are generated and spread under chaotic quantum dynamics. In fermionic systems, non-Gaussianity constitutes an important resource beyond entanglement, characterizing the deviation of generic quantum states from free-fermion states. In this Letter, we propose using the \textit{replica twist entropy}, defined through the expectation value of a replica rotation operator with a tunable angle $α$, as a natural probe of non-Gaussianity in pure fermionic states. After deriving its behavior for typical many-body states, we establish a general framework for analyzing the replica twist entropy in solvable Sachdev-Ye-Kitaev models in the large-$N$ limit. Applying this framework to the dynamics of the thermofield double state, we uncover a novel dynamical phase diagram in the parameter space of $α$ and evolution time $t$, exhibiting spontaneous $Z_2$ symmetry breaking at $α=π/4$ and universal cusps as a function of $α$ in the long-time regime. Our results pave the way for studying non-Gaussianity in strongly correlated fermionic systems and reveal a new class of dynamical transitions probed by non-Gaussianity.

quant-ph↗
arXiv2026

Entanglement Growth from Structured Initial States in Many-Body Localized Systems

Understanding how complex entanglement structures emerge is a central problem in quantum many-body physics. Recent work by Zhang et al. [Phys. Rev. Lett. 137, 020404 (2026)] has considered structured initial states prepared by evolving a product state under a chaotic Hamiltonian for a finite time before quenching to the target Hamiltonian. In this setup, total entanglement entropy growth in many-body localized systems exhibits two distinct regimes, first increasing and then decreasing as the initial entanglement is tuned. In this work, we identify the physical origin of this behavior by analyzing the dynamics of both the Rényi entanglement entropy and the Wehrl-Rényi entropy in the random-field XXZ model, the latter of which characterizes multipartite entanglement. We show that a similar non-monotonic dependence on the initial entanglement also appears in the net growth of the Wehrl-Rényi entropy for product states polarized along the $z$-direction. The first regime is governed by a finite magnetization associated with local integrals of motion, while the second reflects inter-site correlations. In contrast, for product states in the $x/y$-direction, the entanglement growth exhibits a monotonic decay. Our results provide a more fine-grained picture of how distinct initial-state properties shape entanglement dynamics in many-body localized systems.

quant-ph↗
arXiv2026

Few-Magnon Universality beyond Contact Interactions in Long-Range Spin Chains

Quantum few-body systems exhibit a rich variety of phenomena that provide fundamental insights into universality. A paradigmatic example is the Efimov effect, in which an infinite tower of three-body bound states emerges with binding energies forming a geometric sequence. Motivated by advances in trapped-ion systems, recent theoretical work has predicted that the Efimov effect can arise in certain long-range quantum spin chains, where magnons exhibit a general dispersion $ε_k=u|k|^z$ with $z\in(0.52,0.88)$ and interact through a contact potential. Nevertheless, whether Efimov universality persists in the presence of experimentally relevant long-range interactions between magnons remains unexplored. In this work, we address this issue by investigating the two- and three-magnon problems with a model interaction potential $V(x)=V_0/(1+|x|^β)$. Our results demonstrate that few-body universality persists for any $β>1$: In the two-magnon sector, the low-energy expansion of the scattering matrix near a two-body resonance agrees with the universal form predicted by contact-interaction theory. The three-magnon bound-state spectrum exhibits Efimov scaling, with a scaling factor consistent with the theory. We further highlight the experimental relevance of our model, which can be readily realized in trapped-ion systems. Our results pave the way for quantum simulations of universal few-body physics beyond the traditional paradigm of nonrelativistic particles.

cond-mat.quant-gas↗
arXiv2026

PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation

Text-to-audio-video (T2AV) generation is central to applications such as filmmaking and world modeling. However, current models often fail to produce physically plausible sounds. Previous benchmarks primarily focus on audio-video temporal synchronization, while largely overlooking explicit evaluation of audio-physics grounding, thereby limiting the study of physically plausible audio-visual generation. To address this issue, we present PhyAVBench, the first benchmark that systematically evaluates the audio-physics grounding capabilities of T2AV, image-to-audio-video (I2AV), and video-to-audio (V2A) models. PhyAVBench offers PhyAV-Sound-11K, a new dataset of 25.5 hours of 11,605 audible videos collected from 184 participants to ensure diversity and avoid data leakage. It contains 337 paired-prompt groups with controlled physical variations that drive sound differences, each grounded with an average of 17 videos and spanning 6 audio-physics dimensions and 41 fine-grained test points. Each prompt pair is annotated with the physical factors underlying their acoustic differences. Importantly, PhyAVBench leverages paired text prompts to evaluate this capability. We term this evaluation paradigm the Audio-Physics Sensitivity Test (APST) and introduce a novel metric, the Contrastive Physical Response Score (CPRS), which quantifies the acoustic consistency between generated videos and their real-world counterparts. We conduct a comprehensive evaluation of 17 state-of-the-art models. Our results reveal that even leading commercial models struggle with fundamental audio-physical phenomena, exposing a critical gap beyond audio-visual synchronization and pointing to future research directions. We hope PhyAVBench will serve as a foundation for advancing physically grounded audio-visual generation. Prompts, ground-truth, and generated video samples are available at https://github.com/imxtx/PhyAVBench.

cs.SD↗
arXiv2026

ADAPT: Audio-Depth Action Planning with Transition-Aware Rewards for Lights-Out Industrial Navigation

Vision-language-action models for embodied navigation typically rely on RGB observations, limiting their applicability in lights-out industrial environments where visible illumination may be unavailable. We study illumination-independent navigation using spatial audio for target localization and depth for obstacle-aware geometry. To support this setting, we introduce DarkNav-Bench, a benchmark with 24 industrial scenes and four progressively challenging acoustic conditions. We further propose ADAPT, an audio-depth action-planning framework equipped with spatial-audio perception and optimized for closed-loop navigation through reinforcement learning (RL). Our key RL contribution, transition-aware reward shaping (TARS), defines a pose-aware shortest-action potential over a position-heading graph. Unlike conventional geodesic-progress rewards, TARS accounts for both feasible movements and in-place rotations, providing action-consistent feedback for policy optimization. Experiments show that ADAPT consistently outperforms training-based navigation baselines and zero-shot multimodal models on both seen and unseen scenes. On unseen environments, it achieves 90.6% success under Clean conditions and retains 60.0% success under the most challenging Localized Noise setting. These results demonstrate the potential of spatial audio and depth as complementary modalities for reliable RGB-free navigation in lights-out industrial environments.

cs.AI↗
arXiv2026

Cancerous Backdoor Attack: Gradual Activation via Invisible Tokens and A Stage Controller

Backdoor Attacks on Large Language Models (LLMs) typically follow a trigger-response pattern, where satisfying a trigger condition activates a single terminal behavior. However, this single-point backdoor activation does not explore attacks in which malicious behavior develops gradually through multiple intermediate stages before reaching the terminal activation stage. We present Cancerous Backdoor Attack, that gradually corrupts an LLM's reasoning over a sequence of interactions by leveraging invisible tokens and a stage controller. Specifically, invisible tokens provide hidden stages to the model without appearing in the user's visible request and the stage controller tracks the attack's progression during inference to realize gradual activation. We evaluate Cancerous Backdoor Attack across several reasoning datasets. The attack consistently produces gradual reasoning corruption across models and tasks. We compare Cancerous Backdoor Attack with prior backdoor attack baselines to demonstrate our proposed attack outperforms the baselines. Our results indicate that reasoning corruption can develop gradually before it becomes visible, making answer accuracy alone insufficient for detecting backdoor attacks.

cs.CR↗
arXiv2026

Time-Reversal Selection Rule for Twist Disorder

Random layer rotations are ubiquitous in layered matter, from turbostratic films to rotationally disordered crystals, and strongly suppress transport across the layers. Generated by a symmetry operation, this disorder is absent on the rotation axis, so the thickness laws of conduction are set by how fast backscattering vanishes there. We show that a time-reversal selection rule decides this: each twist harmonic of the interlayer bond is either a frame rotation, whose backscattering cancels, or a coupling change, which scatters. A zero-set theorem turns the rule into universal thickness laws for electrons and phonons, fixed by band-edge symmetry; random twist increments rescale the backscattering but keep its order. Where the axis is unprotected, band nodes provide a second route: a transparent energy with heavy-tailed disorder and an $N^{-3}$ law. A first-principles forward-channel model of black phosphorus shows both routes; its random-twist films act as a symmetry filter that, relative to aligned stacks, retains a hundred times more of the electron than of the hole conductance at 64 layers. The symmetry of band edges and phonon branches thus decides what crosses a disordered stack, making stacking disorder a design element.

cond-mat.mtrl-sci↗
arXiv2026

Fermionic Gaussian Scrooge Ensembles in Deep Thermalization

Measuring part of a many-body wave function generates a projected ensemble of pure quantum states on the unmeasured subsystem. Recent advances in deep thermalization have shown that, in chaotic systems, this ensemble universally converges to the maximally random ensemble compatible with its average density matrix, known as the Scrooge ensemble. By contrast, free-fermion systems are nonchaotic, and their quantum states are constrained to remain Gaussian. Motivated by this distinction, we introduce the fermionic Gaussian Scrooge ensemble to describe deep thermalization in generic free-fermion systems. This ensemble is defined as a distortion of the Gaussian Haar ensemble by the average density matrix, and its moments reveal an enlarged symmetry of Gaussian states in the replicated Hilbert space. We demonstrate the emergence of the fermionic Gaussian Scrooge ensemble in two complementary settings: (1) we analytically prove that the projected ensemble of the SYK$_2$ model follows the fermionic Gaussian Scrooge ensemble at arbitrary evolution times; (2) we provide numerical evidence that it emerges at sufficiently long times in random Gaussian circuits with charge conservation. Our results establish the fermionic Gaussian Scrooge ensemble as a universal description of deep thermalization in free-fermion systems.

quant-ph↗
arXiv2026

GUITAR: Structured Failure Diagnosis of GUI Agents via State Transitions

Understanding where and why Graphical User Interface (GUI) agents fail is essential for building more reliable systems, yet current evaluation relies on step accuracy, a metric that treats each screen independently and overlooks the underlying structure of GUI environments. This leads to two critical blind spots: (1) functionally equivalent screens are evaluated in isolation, obscuring systematic failure patterns across shared screens; and (2) the long-tailed GUI distribution renders failures on rare but critical screens invisible under standard metrics. To address these issues, we propose \textbf{GUITAR}, a state-centric diagnostic framework that performs structured failure analysis over both states and transitions, using a State Transition Graph (STG) by mapping visually diverse screens to shared functional states. Across 8 agents and 6 tasks from AndroidControl and Mind2Web, GUITAR reveals that 60.4\% of failures occur in 20\% of states, localizing errors to a small set of bottlenecks. Bottleneck-targeted guidance improves SR by 2.8\% and retains a 1.88\% average gain across 7 agents under three-fold trajectory-held-out evaluation with fully automatic STGs. These findings demonstrate the diagnostic and actionable value of structure-aware evaluation within the evaluated mobile and web tasks. Code is available at https://github.com/sqzhang-lazy/GUITAR

cs.AI↗
arXiv2026

Devils in Question Relay: Source-Conditioned Relay Steering to Mitigate Hallucinations in Audio-visual Large Language Models

Audio-visual large language models (AVLLMs) have made remarkable progress in multimodal understanding and reasoning through interactions among visual, auditory, and linguistic information. However, recent studies show that AVLLMs face a critical challenge: $\textbf{source-confused grounding hallucination}$, where cues from the unused modality induce responses that the required modality does not support, undermining reliability in real-world applications. Existing methods have made progress in mitigating this failure, yet how it arises from internal cross-modal interactions remains insufficiently understood. To address this gap, we conduct path-intervention and representation analyses, revealing a $\textbf{question-relay}$ mechanism: question states carry interfering cues alongside required-source evidence, undermining grounding in required-modality evidence. Cutting pathways from interfering modality to question states yields greater correct-answer logit recovery than cutting those to the generation position. Motivated by these findings, we propose $\textbf{SECRET}$ ($\textbf{S}$ourc$\textbf{E}$-$\textbf{C}$onditioned $\textbf{RE}$lay s$\textbf{T}$eering), a training-free method that mitigates cross-modal interference at the question relay. Using contrasting question representations elicited through different modality-pathway interventions, SECRET steers the original question states toward required-source evidence. Experiments on two widely adopted benchmarks CMM and AVHBench across three AVLLMs show that SECRET consistently outperforms prior training-free methods, substantially mitigating source-confused grounding hallucinations (e.g., up to +18.0 and +7.1 percentage points over base models). Modality-specific captioning further demonstrates its generalizability to open-ended generation.

cs.CL↗
arXiv2026

SEmoEdit: Probing and Harnessing the Editability of Pre-trained Speech Flows

Existing training-based speech emotion editing methods often require substantial task-specific training and can be unstable. This motivates us to investigate whether the pretrained generative dynamics of large-scale text-to-speech (TTS) models can be directly manipulated for training-free emotion editing. To answer this question, we probe the editability of pretrained flow-matching and hybrid TTS models by constructing a controlled test set and systematically diagnosing editing effects along the generative trajectory. Our analysis reveals that pretrained TTS models are substantially editable in emotion, but such editability is architecture- and trajectory-dependent and can be disrupted by early flow-matching steps, while cross-speaker emotion transport carries additional acoustic attributes beyond emotion. To address these limitations, we propose SEmoEdit, the first training-free framework that formulates emotion editing as dynamic velocity transport between source and target emotions, enabling robust, flow-based speech emotion editing directly within pretrained TTS models. SEmoEdit unifies three core operations: emotion replacement, emotion erasure, and continuous emotion interpolation, requiring neither parameter updates nor task-specific optimization. To systematically evaluate these capabilities, we introduce SEmoEditBench, a dataset comprising 600 editing cases, and conduct extensive experiments across state-of-the-art (SOTA) models and backbones. Our results show that SEmoEdit is highly effective and broadly applicable, outperforming existing training-based and activation-steering methods. Ultimately, this work reveals that pretrained speech flows possess rich, latent emotion-editing capabilities, providing useful guidance for real applications. Code, benchmark, and Audio samples are available at https://github.com/imxtx/SEmoEdit.

cs.SD↗
arXiv2026

The Platonic brain bridge hypothesis: human brain networks as an architectural prior for multimodal large language models

Multimodal large language models predict brain activity, but brain alignment has been a measurement, not a design tool. We propose the Platonic brain bridge hypothesis: omni models, multimodal large language models that process video, audio and text jointly, converge on brain-like representations usable in both directions. From model to brain, brain-likeness of seven omni models is stable across participants, rises with every input channel in three bases, and our encoders lead the Algonauts 2025 out-of-distribution leaderboard. From brain to model, Brain-MoE fixes the expert partition of a frozen base to the seven networks of human cortex, trains experts on network-labelled Brain-AVQA questions, raises held-out accuracy in all 15 model-benchmark pairs by 6.42 percentage points on average and exceeds capacity-matched random experts in 14. Brain-Scope localizes the correspondence to sparse features whose removal weakens brain prediction. Human brain organization is therefore a usable architectural prior for multimodal large language models.

q-bio.NC↗
arXiv2026

Proving olympiad geometry theorems on a superconducting quantum processor

Automated theorem proving seeks to use computational systems to prove or disprove mathematical and logical statements [1, 2]. It underpins a wide range of applications, and enhancing theorem-proving capabilities remains a central objective in artificial intelligence [3]. Although recent neuro-symbolic systems have achieved remarkable progress [4-7], their operation is ultimately constrained by classical computational architectures. Quantum computing [8], by contrast, enables information encoding and coherent parallelism beyond classical limits [9-14], raising the possibility of accelerating structured symbolic deduction [15]. Here we report the experimental realization of automated geometry theorem proving on a fully programmable superconducting quantum processor. We develop two complementary quantum proving frameworks. The first implements Wu's algebraic elimination method using quantum pseudo-division, with multivariate polynomials represented in superposition states, enabling quantum algebraic theorem proving. The second implements the full-angle method as backward symbolic reasoning through a hybrid quantum strategy-guided architecture, demonstrating a general route toward quantum symbolic proof search. As illustrative examples, we prove two theorems on a superconducting quantum processor: the perpendicularity of the diagonals of a square and a 1978 International Mathematical Olympiad geometry problem. Our results establish, at the experimental level, automated logical reasoning as a viable task for near-term quantum processors and provide a concrete pathway toward quantum-enhanced symbolic intelligence.

quant-ph↗
arXiv2026

Predicting energy and structural response to force correction in molecular dynamics

We predict how force correction changes energy exchange and structural statistics by measuring leading response coefficients on shared reference trajectories. Residual power and the displacement virial distinguish the transfer of energy from the change in restoring forces, including intermittent reference updates. Independent simulations then test the predicted kinetic and configurational shifts. A matched-timetable experiment shows that reference timing affects heating through its coupling to the evolving state. In an anharmonic chain, the displacement virial predicts a structural shift missed by a power-only description. Local force constants in silicon predict a complementary directional tradeoff: scalar calibration repairs optical motion while degrading an already accurate low-frequency direction. Full nonlinear trajectories confirm this tradeoff and distinguish the benefits of static curvature correction and repeated reference impulses. Independent finite-temperature integrals in orthorhombic tin selenide (SnSe) support the configurational-response direction predicted from separate reference calculations. These results provide a physical basis for choosing how reference information enters molecular dynamics. The framework assesses force correction through its effect on atomic motion and statistical observables, beyond the accuracy of individual force evaluations.

cond-mat.mtrl-sci↗
arXiv2026

Beyond Noise Steering: Dual-Latent Space Reinforcement Learning for Generative Robot Policy

Pretrained generative robot policies learn expressive action priors from demonstrations. However, existing reinforcement learning methods only steer the noisy space but fail to modulate intermediate action representations during the generation process, resulting in performance degradation and inefficiency. To address this limitation, we propose a novel Dual-Latent Space Reinforcement Learning (DLSRL) framework, which complements initial-noise steering with representation-level control inside the frozen generator. Specifically, our actor network predicts two distinct latent variables: an initial-noise latent variable that steers behavior generation, and an action-representation latent variable for intermediate feature modulation. Moreover, this representation latent variable is mapped to adapter features and ingeniously injected into the hidden states of intermediate action tokens via residual connections. Our dual-control design enables direct adjustment of action representations without updating the base policy. Experiments across generative policy architectures and robotic manipulation tasks show that DLSRL effectively accelerates online robot policy adaptation and achieves competitive performance. Our code is available at \href{https://github.com/xianchaoxiu/DLSRL}{https://github.com/xianchaoxiu/DLSRL}.

cs.RO↗
arXiv2026

Do Video Generators Track the World Across Segments? A Benchmark and Method for World-State Reasoning in Video Continuation

Video generators build long videos by composing shorter parts, either by generating segments one after another or by autoregressively extending chunks. Each new part usually depends on memories of historical observations, such as recent frames, selected key frames, memory banks, or cached features. These memories preserve visible evidence from the past, but current generators do not reliably turn such evidence into a world-state interface: what holds in the video world after previous actions and how it should change under the next prompt. A past frame remains valid history, but it may not describe the state needed by the next segment; some states must instead be inferred from occluded or implicit changes rather than copied from a directly observed frame. This creates a simple but overlooked question for video continuation: given a previous video, its prompt, and a new prompt, can a model generate a continuation that reflects the state determined by both the historical video and the new prompt? To answer this question, we introduce Statebench, a benchmark that targets this gap by testing continuations over three state categories: past-visible states, occluded-process states, and complex-transition states. We further propose Stateagent, which explicitly maintains an entity-state representation, updates it under the new prompt, grounds the predicted post-action state as a future end frame, and renders the next video. Experiments show that our method improves controlled video continuation by raising the all-case state score (SCS-All) from 45.2 to 69.3, and also benefits story generation at the one-minute scale. Code is avaliable at https://github.com/AMAP-ML/StateAgent.

cs.CV↗
arXiv2026

Reconstruction of anomalous air showers with SKA-Low

Double-bump showers are a surprising class of extensive air showers (EAS) predicted by Monte Carlo simulations, which, so far, no experiment has been able to directly detect. They occur when a high-energy secondary particle, the leading particle, travels significantly farther than the rest, creating a distinct double-peaked longitudinal profile. The unique radio footprint of double-bump showers, characterized by multiple pulses in the signals and interference patterns in the frequency spectra, enables reconstruction of longitudinal profiles from radio observations. With its dense antenna array and broad frequency range, SKA-Low will be the first observatory capable of detecting these features, offering a new opportunity to probe hadronic interactions and use the distinctive signatures of elements to provide new mass composition measurements. The goal of this analysis is to take the first steps toward using these radio signatures to reconstruct the relevant parameters of the longitudinal profile of a double-bump shower. We will start by explaining the radio signal of double-bump}showers compared to that of average showers. Then we will create a simple 2-point emission model to explain the interference patterns in the frequency spectra, which can be inverted to obtain rudimentary estimates of atmospheric depth of both peaks. Lastly, we implement a brute-force approach to reconstruct multiple parameters of the double bump.

astro-ph.HE↗