arXiv subjects

Jie Gu

Publications and source records attributed to Jie Gu.

At least 37 records · Page 2Linked to original sources

arXiv2026

Geometric Floquet Condition for Quantum Adiabaticity

Quantum adiabaticity is the evolution of a quantum system that remains close to an instantaneous eigenstate of a time-dependent Hamiltonian. Using Floquet formalism, we derive a rigorous sufficient condition for adiabaticity in closed, finite-dimensional periodically driven systems that is valid for arbitrarily many driving periods. The condition is stroboscopic and geometric, depending only on single-cycle information: the Fubini--Study length of the instantaneous eigenray and a quasienergy-separation measure extracted from the Floquet operator. We also formulate a state-targeted refinement that reduces conservativeness when only one adiabatic branch is relevant. Rather than synthesizing control pulses, the result provides a certification criterion for a given periodic protocol. We illustrate the criterion and contrast it with conventional instantaneous-gap conditions in three representative examples.

quant-ph↗

arXiv2026

Beyond Hagedorn: A Harmonic Approach to $T\bar{T}$-deformation

We apply harmonic analysis to study the $T\bar{T}$-deformed torus partition function. We first express the CFT partition functions in terms of Maass waveforms, including the Eisenstein series and cusp forms. These basis functions turn out to deform in a very simple way under the $T\bar{T}$-deformation. The spectral decomposition provides a numerically stable and efficient method to compute the partition function at finite values of the deformation parameter $λ$, allowing us to clearly resolve the analytic structure of the partition function as a function of $λ$. The resulting deformed partition function exhibits a Hagedorn singularity. Building on harmonic analysis approach, we propose a natural analytic continuation beyond the Hagedorn singularity, which enables us to compute the full partition function for any value of $λ$.

hep-th↗

arXiv2026

Task-Related Token Compression in Multimodal Large Language Models from an Explainability Perspective

Existing Multimodal Large Language Models (MLLMs) process a large number of visual tokens, leading to significant computational costs and inefficiency. Instruction-related visual token compression demonstrates strong task relevance, which aligns well with MLLMs ultimate goal of instruction following. Previous works generally assume that visual tokens achieve better vision-language alignment in the shallow layers of LLMs, which have led to task-related token compression being primarily applied in intermediate LLM layers. In contrast, our study reveals that with proper selection, task-related token compression is feasible at the input stage of LLM with negligible performance loss. This new paradigm significantly reduces task-irrelevant visual tokens and its model-agnostic design enables application without modifying the LLM architecture. Specifically, we suggest that explainability methods for transformer-based architechtures can evaluate the global importance of each visual token with respect to the given instruction, which can effectively guide the task-related token compression for MLLMs. Furthermore, we propose to learn a mapping from the attention map of the first LLM layer to the explanation results, thereby avoiding the need for a full inference pass. Interestingly, this mapping can be learned using a simple and lightweight convolutional network, whose training is efficient and independent of MLLMs. Extensive experiments on 13 image and video benchmarks across three leading MLLMs (Qwen2-VL, LLaVA-OneVision, and VILA1.5) demonstrate the remarkable effectiveness and strong generalization of our approach. Additionally, our new compression paradigm achieves faster inference with reductions in both prefilling time and KV cache memory.

cs.CV↗

arXiv2026

Spectral Fluctuation-Dissipation-Response Inequalities

We derive spectral fluctuation--dissipation--response inequalities for finite-state Markov jump processes. By comparing the causal susceptibility to its passive equilibrium reference, we establish frequency-resolved and frequency-integrated inequalities that bound their mismatch in terms of the steady-state entropy production rate, probe variance, short-time perturbation diffusion, and reversible relaxation timescales. Our bounds exactly recover the standard fluctuation--dissipation theorem at equilibrium and apply directly to measurable causal susceptibilities, providing experimentally testable thermodynamic limits on FDT breakdown in driven steady states.

cond-mat.stat-mech↗

arXiv2026

Comment on "Extension of the adiabatic theorem"

Phys. Rev. B 113, 165102 (2026) proposed the conjecture that, for quantum quenches within the same phase, the overlap between the initial ground state and postquench eigenstates is maximal for the postquench ground state. We show that this conjecture is not valid in general. An explicit local, translationally invariant, gapped free-fermion counterexample exists even though the pre- and postquench Hamiltonians are connected by a symmetry-preserving gapped path and the thermodynamic-limit spectrum is continuous.

cond-mat.stat-mech↗

arXiv2026

Resurgence of Chern-Simons theory at the trivial flat connection

Some years ago, it was conjectured by the first author that the Chern-Simons perturbation theory of a 3-manifold at the trivial flat connection is a resurgent power series. We describe completely the resurgent structure of the above series (including the location of the singularities and their Stokes constants) in the case of a hyperbolic knot complement in terms of an extended square matrix of $(x,q)$-series whose rows are indexed by the boundary parabolic $\text{SL}_2(\mathbb{C})$-flat connections, including the trivial one. We use our extended matrix to describe the Stokes constants of the above series, to define explicitly their Borel transform and to identify it with state-integrals. Along the way, we use our matrix to give an analytic extension of the Kashaev invariant and of the colored Jones polynomial and to complete the matrix valued holomorphic quantum modular forms as well as to give an exact version of the refined quantum modularity conjecture of Zagier and the first author. Finally, our matrix provides an extension of the 3D-index in a sector of the trivial flat connection. We illustrate our definitions, theorems, numerical calculations and conjectures with the two simplest hyperbolic knots.

math.GT↗

arXiv2026

RhoMorph: Rhombus-shaped Deformable Modular Robots for Stable, Medium-Independent Reconfiguration Motion

In this paper, we present RhoMorph, a novel deformable planar lattice modular self-reconfigurable robot (MSRR) with a rhombus shaped module. Each module consists of a parallelogram skeleton with a single centrally mounted actuator that enables folding and unfolding along its diagonal. The core design philosophy is to achieve essential MSRR functionalities such as morphing, docking, and locomotion with minimal control complexity. This enables a continuous and stable reconfiguration process that is independent of the surrounding medium, allowing the system to reliably form various configurations in diverse environments. To leverage the unique kinematics of RhoMorph, we introduce morphpivoting, a novel motion primitive for reconfiguration that differs from advanced MSRR systems, and propose a strategy for its continuous execution. Finally, a series of physical experiments validate the module's stable reconfiguration ability, as well as its positional and docking accuracy.

cs.RO↗

arXiv2026

Resurgent structure of 2d Yang-Mills theory on a torus

We study the resurgent structure of the topological string dual to 2d $U(N)$ Yang-Mills on torus. We find closed form formulas for instanton amplitudes up to arbitrarily high instanton orders, based on which we propose the non-perturbative partition function including contributions from all the real instantons, which is real for positive modulus and string coupling. We also explore complex instantons and find two infinite towers of them. We expect them to correspond to BPS states in type II string.

hep-th↗

arXiv2026

Towards Dexterous Embodied Manipulation via Deep Multi-Sensory Fusion and Sparse Expert Scaling

Realizing dexterous embodied manipulation necessitates the deep integration of heterogeneous multimodal sensory inputs. However, current vision-centric paradigms often overlook the critical force and geometric feedback essential for complex tasks. This paper presents DeMUSE, a Deep Multimodal Unified Sparse Experts framework leveraging a Diffusion Transformer to integrate RGB, depth, and 6-axis force into a unified serialized stream. Adaptive Modality-specific Normalization (AdaMN) is employed to recalibrate modality-aware features, mitigating representation imbalance and harmonizing the heterogeneous distributions of multi-sensory signals. To facilitate efficient scaling, the architecture utilizes a Sparse Mixture-of-Experts (MoE) with shared experts, increasing model capacity for physical priors while maintaining the low inference latency required for real-time control. A Joint denoising objective synchronously synthesizes environmental evolution and action sequences to ensure physical consistency. Achieving success rates of 83.2% and 72.5% in simulation and real-world trials, DeMUSE demonstrates state-of-the-art performance, validating the necessity of deep multi-sensory integration for complex physical interactions.

cs.RO↗

arXiv2026

Gaussian Sequences with Multi-Scale Dynamics for 4D Reconstruction from Monocular Casual Videos

Understanding dynamic scenes from casual videos is critical for scalable robot learning, yet four-dimensional (4D) reconstruction under strictly monocular settings remains highly ill-posed. To address this challenge, our key insight is that real-world dynamics exhibits a multi-scale regularity from object to particle level. To this end, we design the multi-scale dynamics mechanism that factorizes complex motion fields. Within this formulation, we propose Gaussian sequences with multi-scale dynamics, a novel representation for dynamic 3D Gaussians derived through compositions of multi-level motion. This layered structure substantially alleviates ambiguity of reconstruction and promotes physically plausible dynamics. We further incorporate multi-modal priors from vision foundation models to establish complementary supervision, constraining the solution space and improving the reconstruction fidelity. Our approach enables accurate and globally consistent 4D reconstruction from monocular casual videos. Experiments of dynamic novel-view synthesis (NVS) on benchmark and real-world manipulation datasets demonstrate considerable improvements over existing methods.

cs.CV↗

arXiv2026

Self-Reconfiguration Planning for Deformable Quadrilateral Modular Robots

For lattice modular self-reconfigurable robots (MSRRs), maintaining stable connections during reconfiguration is crucial for physical feasibility and deployability. This letter presents a novel self-reconfiguration planning algorithm for deformable quadrilateral MSRRs that guarantees stable connection. The method first constructs feasible connect/disconnect actions using a virtual graph representation, and then organizes these actions into a valid execution sequence through a Dependence-based Reverse Tree (DRTree) that resolves interdependencies. We also prove that reconfiguration sequences satisfying motion characteristics exist for any pair of configurations with seven or more modules (excluding linear topologies). Finally, comparisons with a modified BiRRT algorithm highlight the superior efficiency and stability of our approach, while deployment on a physical robotic platform confirms its practical feasibility.

cs.RO↗

arXiv2026

Counterexamples to the conjectured ordering between the waiting-time bound and the thermodynamic uncertainty bound on entropy production

Two widely used model-free lower bounds on the steady-state entropy production rate of a continuous-time Markov jump process are the thermodynamic uncertainty relation (TUR) bound $σ_\text{TUR}$, derived from the mean and variance of a current, and the waiting-time distribution (WTD) bound $σ_\mathcal{L}$, derived from the irreversibility of partially observed transition sequences together with their waiting times. It has been conjectured that $σ_{\mathcal L}$ is always at least as tight as $σ_{\mathrm{TUR}}$ when both are constructed from the same partially observed link. Here we provide four-state counterexamples in a nonequilibrium steady state where $σ_{\mathcal L}<σ_{\mathrm{TUR}}$. This result shows that no universal ordering exists between these two inference bounds under partial observation.

cond-mat.stat-mech↗

arXiv2025

LLA: Enhancing Security and Privacy for Generative Models with Logic-Locked Accelerators

We introduce LLA, an effective intellectual property (IP) protection scheme for generative AI models. LLA leverages the synergy between hardware and software to defend against various supply chain threats, including model theft, model corruption, and information leakage. On the software side, it embeds key bits into neurons that can trigger outliers to degrade performance and applies invariance transformations to obscure the key values. On the hardware side, it integrates a lightweight locking module into the AI accelerator while maintaining compatibility with various dataflow patterns and toolchains. An accelerator with a pre-stored secret key acts as a license to access the model services provided by the IP owner. The evaluation results show that LLA can withstand a broad range of oracle-guided key optimization attacks, while incurring a minimal computational overhead of less than 0.1% for 7,168 key bits.

cs.CR↗

arXiv2025

Marginalize, Rather than Impute: Probabilistic Wind Power Forecasting with Incomplete Data

Machine learning methods are widely and successfully used for probabilistic wind power forecasting, yet the pervasive issue of missing values (e.g., due to sensor faults or communication outages) has received limited attention. The prevailing practice is impute-then-predict, but conditioning on point imputations biases parameter estimates and fails to propagate uncertainty from missing features. Our approach treats missing features and forecast targets uniformly: we learn a joint generative model of features and targets from incomplete data and, at operational deployment, condition on the observed features and marginalize the unobserved ones to produce forecasts. This imputation-free procedure avoids error introduced by imputation and preserves uncertainty aroused from missing features. In experiments, it improves forecast quality in terms of continuous ranked probability score relative to impute-then-predict baselines while incurring substantially lower computational cost than common alternatives.

cs.LG↗

arXiv2025

MODUR: A Modular Dual-reconfigurable Robot

Modular Self-Reconfigurable Robot (MSRR) systems are a class of robots capable of forming higher-level robotic systems by altering the topological relationships between modules, offering enhanced adaptability and robustness in various environments. This paper presents a novel MSRR called MODUR, featuring dual-level reconfiguration capabilities designed to integrate reconfigurable mechanisms into MSRR. Specifically, MODUR can perform high-level self-reconfiguration among modules to create different configurations, while each module is also able to change its shape to execute basic motions. The design of MODUR primarily includes a compact connector and scissor linkage groups that provide actuation, forming a parallel mechanism capable of achieving both connector motion decoupling and adjacent position migration capabilities. Furthermore, the workspace, considering the interdependent connectors, is comprehensively analyzed, laying a theoretical foundation for the design of the module's basic motion. Finally, the motion of MODUR is validated through a series of experiments.

cs.RO↗

arXiv2025

Egocentric Instruction-oriented Affordance Prediction via Large Multimodal Model

Affordance is crucial for intelligent robots in the context of object manipulation. In this paper, we argue that affordance should be task-/instruction-dependent, which is overlooked by many previous works. That is, different instructions can lead to different manipulation regions and directions even for the same object. According to this observation, we present a new dataset comprising fifteen thousand object-instruction-affordance triplets. All scenes in the dataset are from an egocentric viewpoint, designed to approximate the perspective of a human-like robot. Furthermore, we investigate how to enable large multimodal models (LMMs) to serve as affordance predictors by implementing a ``search against verifiers'' pipeline. An LMM is asked to progressively predict affordances, with the output at each step being verified by itself during the iterative process, imitating a reasoning process. Experiments show that our method not only unlocks new instruction-oriented affordance prediction capabilities, but also achieves outstanding performance broadly.

cs.RO↗

arXiv2025

Stimulating Imagination: Towards General-purpose "Something Something Placement"

General-purpose object placement is a fundamental capability of an intelligent generalist robot: being capable of rearranging objects following precise human instructions even in novel environments. This work is dedicated to achieving general-purpose object placement with ``something something'' instructions. Specifically, we break the entire process down into three parts, including object localization, goal imagination and robot control, and propose a method named SPORT. SPORT leverages a pre-trained large vision model for broad semantic reasoning about objects, and learns a diffusion-based pose estimator to ensure physically-realistic results in 3D space. Only object types (movable or reference) are communicated between these two parts, which brings two benefits. One is that we can fully leverage the powerful ability of open-set object recognition and localization since no specific fine-tuning is needed for the robotic scenario. Moreover, the diffusion-based estimator only need to ``imagine" the object poses after the placement, while no necessity for their semantic information. Thus the training burden is greatly reduced and no massive training is required. The training data for the goal pose estimation is collected in simulation and annotated by using GPT-4. Experimental results demonstrate the effectiveness of our approach. SPORT can not only generate promising 3D goal poses for unseen simulated objects, but also be seamlessly applied to real-world settings.

cs.RO↗

arXiv2025

Resurgent Wilson loops in refined topological string

We study the resurgent structures of Wilson loops in refined topological string theory. We argue that the Borel singularities should be integral periods, and that the associated Stokes constants are refined Donaldson-Thomas invariants, just like the free energies, except that the Borel singularities cannot be local flat coordinates. We also solve the non-perturbative series in closed form from the holomorphic anomaly equations for the refined Wilson loops. We illustrate these results with the examples of local P^2 and local P^1 x P^1.

hep-th↗