arXiv subjects

Jie Zhao

Publications and source records attributed to Jie Zhao.

At least 55 records · Page 3Linked to original sources

arXiv2026

DM0: An Embodied-Native Vision-Language-Action Model towards Physical AI

Moving beyond the traditional paradigm of adapting internet-pretrained models to physical tasks, we present DM0, an Embodied-Native Vision-Language-Action (VLA) framework designed for Physical AI. Unlike approaches that treat physical grounding as a fine-tuning afterthought, DM0 unifies embodied manipulation and navigation by learning from heterogeneous data sources from the onset. Our methodology follows a comprehensive three-stage pipeline: Pretraining, Mid-Training, and Post-Training. First, we conduct large-scale unified pretraining on the Vision-Language Model (VLM) using diverse corpora--seamlessly integrating web text, autonomous driving scenarios, and embodied interaction logs-to jointly acquire semantic knowledge and physical priors. Subsequently, we build a flow-matching action expert atop the VLM. To reconcile high-level reasoning with low-level control, DM0 employs a hybrid training strategy: for embodied data, gradients from the action expert are not backpropagated to the VLM to preserve generalized representations, while the VLM remains trainable on non-embodied data. Furthermore, we introduce an Embodied Spatial Scaffolding strategy to construct spatial Chain-of-Thought (CoT) reasoning, effectively constraining the action solution space. Experiments on the RoboChallenge benchmark demonstrate that DM0 achieves state-of-the-art performance in both Specialist and Generalist settings on Table30.

cs.RO↗

arXiv2026

V2P-Bench: Evaluating Video-Language Understanding with Visual Prompts for Better Human-Model Interaction

Large Vision-Language Models (LVLMs) have made significant strides in the field of video understanding in recent times. Nevertheless, existing video benchmarks predominantly rely on text prompts for evaluation, which often require complex referential language and diminish both the accuracy and efficiency of human model interaction in turn. To address this limitation, we propose V2P-Bench, a robust and comprehensive benchmark for evaluating the ability of LVLMs to understand Video Visual Prompts in human model interaction scenarios. V2P-Bench consists of 980 videos and 1172 well-structured high-quality QA pairs, each paired with manually annotated visual prompt frames. The benchmark spans three main tasks and twelve categories, thereby enabling fine-grained, instance-level evaluation. Through an in-depth analysis of current LVLMs, we identify several key findings: 1) Visual prompts are both more model-friendly and user-friendly in interactive scenarios than text prompts, leading to significantly improved model performance and enhanced user experience. 2) Models are reasonably capable of zero-shot understanding of visual prompts, but struggle with spatiotemporal understanding. Even o1 achieves only 71.8%, far below the human expert score of 88.3%, while most open-source models perform below 60%. 3) LVLMs exhibit pervasive Hack Phenomena in video question answering tasks, which become more pronounced as video length increases and frame sampling density decreases, thereby inflating performance scores artificially. We anticipate that V2P-Bench will not only shed light on these challenges but also serve as a foundational tool for advancing human model interaction and improving the evaluation of video understanding.

cs.CV↗

arXiv2026

Capacity-Achieving Entanglement Purification Protocol for Pauli Dephasing Channel

Quantum communication enables secure information transmission and entanglement distribution, but these tasks are fundamentally limited by the capacities of quantum channels. While quantum repeaters can mitigate losses and noise, entanglement swapping via a central node is ineffective against the Pauli dephasing channel due to degradation from Bell-state measurements. This suggests that purifying distributed Bell states before entanglement swapping is necessary. Although one-way hashing codes are known to saturate the dephasing channel capacity, no explicit two-way purification protocol has previously been shown to achieve this bound. In this work, we present a two-way entanglement purification protocol with an explicit, scalable circuit that asymptotically achieves the dephasing channel capacity. With each iteration, the fidelity of Bell states increases. At the final round, the residual dephasing error is suppressed doubly-exponentially, scaling as $\mathcal{O}(p^{2^{n}})$, enabling near-perfect Bell pairs for any fixed number of purification rounds $n$. The explicit circuit we propose is versatile and applicable to any number of Bell pairs, offering a practical solution for mitigating decoherence in quantum networks and distributed.

quant-ph↗

arXiv2026

Dual-pronged deep learning preprocessing on heterogeneous platforms with CPU, Accelerator and CSD

For image-related deep learning tasks, the first step often involves reading data from external storage and performing preprocessing on the CPU. As accelerator speed increases and the number of single compute node accelerators increases, the computing and data transfer capabilities gap between accelerators and CPUs gradually increases. Data reading and preprocessing become progressively the bottleneck of these tasks. Our work, DDLP, addresses the data computing and transfer bottleneck of deep learning preprocessing using Computable Storage Devices (CSDs). DDLP allows the CPU and CSD to efficiently parallelize preprocessing from both ends of the datasets, respectively. To this end, we propose two adaptive dynamic selection strategies to make DDLP control the accelerator to automatically read data from different sources. The two strategies trade-off between consistency and efficiency. DDLP achieves sufficient computational overlap between CSD data preprocessing and CPU preprocessing, accelerator computation, and accelerator data reading. In addition, DDLP leverages direct storage technology to enable efficient SSD-to-accelerator data transfer. In addition, DDLP reduces the use of expensive CPU and DRAM resources with more energy-efficient CSDs, alleviating preprocessing bottlenecks while significantly reducing power consumption. Extensive experimental results show that DDLP can improve learning speed by up to 23.5% on ImageNet Dataset while reducing energy consumption by 19.7% and CPU and DRAM usage by 37.6%. DDLP also improves the learning speed by up to 27.6% on the Cifar-10 dataset.

cs.DC↗

arXiv2026

WorldVQA: Measuring Atomic World Knowledge in Multimodal Large Language Models

We introduce WorldVQA, a benchmark designed to evaluate the atomic visual world knowledge of Multimodal Large Language Models (MLLMs). Unlike current evaluations, which often conflate visual knowledge retrieval with reasoning, WorldVQA decouples these capabilities to strictly measure "what the model memorizes." The benchmark assesses the atomic capability of grounding and naming visual entities across a stratified taxonomy, spanning from common head-class objects to long-tail rarities. We expect WorldVQA to serve as a rigorous test for visual factuality, thereby establishing a standard for assessing the encyclopedic breadth and hallucination rates of current and next-generation frontier models.

cs.CV↗

arXiv2026

Coherent Amplifier-Empowered Quantum Interferometer: Preserving Sensitivity and Quantum Advantage under High Loss

Quantum interferometers offer phase measurement capabilities that surpass the standard quantum limit (SQL), with phase sensitivity and quantum enhancement factor serving as key performance metrics. However, practical implementations face severe degradation of both metrics due to unavoidable losses, representing the foremost challenge in advancing quantum interferometry toward real-world applications. To address this challenge, we propose a coherent-amplifier-empowered quantum interferometer. The coherent amplifier dramatically suppresses the decay of both sensitivity and quantum enhancement under high-loss conditions, maintaining phase sensitivity beyond the original SQL even for losses exceeding 90%. Using an injected 4.2 dB squeezed-vacuum state in experimental demonstration, our scheme reduces the quantum enhancement degradation under 90% loss from 3.7 dB in a conventional quantum interferometer (CQI) to only 1.5 dB. More importantly, the phase sensitivity degradation under the same loss is limited to 4.0 dB, markedly outperforming the 11.2 dB degradation observed in a CQI. This improvement is enabled by the coherent amplifier's phase-sensitive photon amplification and its protection of the quantum state. This breakthrough in amplifier-empowered quantum interferometry overcomes the critical barrier to practical deployment, enabling robust quantum-enhanced measurements in lossy environments.

quant-ph↗

arXiv2026

Precision Bounds for Characterising Quantum Measurements

Quantum measurements, alongside quantum states and processes, form a cornerstone of quantum information processing. However, unlike states and processes, their efficient characterisation remains relatively unexplored. We resolve this asymmetry by introducing a comprehensive framework for efficient detector estimation that reveals the fundamental limits to extractable parameter information and errors arising in detector analysis - the detector quantum Fisher information. Our development eliminates the need to optimise for the best probe state, while highlighting aspects of detector analysis that fundamentally differ from quantum state estimation. Through proofs, examples and experimental validation, we demonstrate the relevance and robustness of our proposal for current quantum detector technologies. By formalising a dual perspective to state estimation, our framework completes and connects the triad of efficient state, process, and detector tomography, advancing quantum information theory with broader implications for emerging technologies reliant on precisely calibrated measurements.

quant-ph↗

arXiv2026

FastStair: Learning to Run Up Stairs with Humanoid Robots

Running up stairs is effortless for humans but remains extremely challenging for humanoid robots due to the simultaneous requirements of high agility and strict stability. Model-free reinforcement learning (RL) can generate dynamic locomotion, yet implicit stability rewards and heavy reliance on task-specific reward shaping tend to result in unsafe behaviors, especially on stairs; conversely, model-based foothold planners encode contact feasibility and stability structure, but enforcing their hard constraints often induces conservative motion that limits speed. We present FastStair, a planner-guided, multi-stage learning framework that reconciles these complementary strengths to achieve fast and stable stair ascent. FastStair integrates a parallel model-based foothold planner into the RL training loop to bias exploration toward dynamically feasible contacts and to pretrain a safety-focused base policy. To mitigate planner-induced conservatism and the discrepancy between low- and high-speed action distributions, the base policy was fine-tuned into speed-specialized experts and then integrated via Low-Rank Adaptation (LoRA) to enable smooth operation across the full commanded-speed range. We deploy the resulting controller on the Oli humanoid robot, achieving stable stair ascent at commanded speeds up to 1.65 m/s and traversing a 33-step spiral staircase (17 cm rise per step) in 12 s, demonstrating robust high-speed performance on long staircases. Notably, the proposed approach served as the champion solution in the Canton Tower Robot Run Up Competition.

cs.RO↗

arXiv2025

AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis

Modern AI models demand high-performance computation kernels. The growing complexity of LLMs, multimodal architectures, and recommendation systems, combined with techniques like sparsity and quantization, creates significant computational challenges. Moreover, frequent hardware updates and diverse chip architectures further complicate this landscape, requiring tailored kernel implementations for each platform. However, manual optimization cannot keep pace with these demands, creating a critical bottleneck in AI system development. Recent advances in LLM code generation capabilities have opened new possibilities for automating kernel development. In this work, we propose AKG kernel agent (AI-driven Kernel Generator), a multi-agent system that automates kernel generation, migration, and performance tuning. AKG kernel agent is designed to support multiple domain-specific languages (DSLs), including Triton, TileLang, CPP, and CUDA-C, enabling it to target different hardware backends while maintaining correctness and portability. The system's modular design allows rapid integration of new DSLs and hardware targets. When evaluated on KernelBench using Triton DSL across GPU and NPU backends, AKG kernel agent achieves an average speedup of 1.46$\times$ over PyTorch Eager baselines implementations, demonstrating its effectiveness in accelerating kernel development for modern AI workloads.

cs.AI↗

arXiv2025

Information-Scrambling-Enhanced Quantum Sensing Beyond the Standard Quantum Limit

Quantum sensing promises measurement precision beyond classical limits, but its practical realization is often hindered by decoherence and the challenges of generating and stabilizing entanglement in large-scale systems. Here, we experimentally demonstrate a scalable, scrambling-enhanced quantum sensing protocol, referred to as butterfly metrology, implemented on a cross-shaped superconducting quantum processor. By harnessing quantum information scrambling, the protocol converts local interactions into delocalized metrologically useful correlations, enabling robust signal amplification through interference of the scrambled and polarized quantum states. We validate the time-reversal ability via Loschmidt echo measurements and quantify the information scrambling through out-of-time-ordered correlators, establishing the essential quantum resources of our protocol. Our measurements reveal that the sensing sensitivity surpasses the standard quantum limit (SQL) with increasing qubit number, reaching 3.78 in a 9-qubit configuration, compared to the SQL of 3.0. The scheme further exhibits inherent robustness to coherent control errors and probed signal noise. This work demonstrates a readily scalable path toward practical quantum sensing advantages with prevalent experimental platforms.

quant-ph↗

arXiv2025

DiffPixelFormer: Differential Pixel-Aware Transformer for RGB-D Indoor Scene Segmentation

Indoor semantic segmentation is fundamental to computer vision and robotics, supporting applications such as autonomous navigation, augmented reality, and smart environments. Although RGB-D fusion leverages complementary appearance and geometric cues, existing methods often depend on computationally intensive cross-attention mechanisms and insufficiently model intra- and inter-modal feature relationships, resulting in imprecise feature alignment and limited discriminative representation. To address these challenges, we propose DiffPixelFormer, a differential pixel-aware Transformer for RGB-D indoor scene segmentation that simultaneously enhances intra-modal representations and models inter-modal interactions. At its core, the Intra-Inter Modal Interaction Block (IIMIB) captures intra-modal long-range dependencies via self-attention and models inter-modal interactions with the Differential-Shared Inter-Modal (DSIM) module to disentangle modality-specific and shared cues, enabling fine-grained, pixel-level cross-modal alignment. Furthermore, a dynamic fusion strategy balances modality contributions and fully exploits RGB-D information according to scene characteristics. Extensive experiments on the SUN RGB-D and NYUDv2 benchmarks demonstrate that DiffPixelFormer-L achieves mIoU scores of 54.28% and 59.95%, outperforming DFormer-L by 1.78% and 2.75%, respectively. Code is available at https://github.com/gongyan1/DiffPixelFormer.

cs.CV↗

arXiv2025

Tunable Hybrid-Mode Coupler Enabling Strong Interactions between Transmons at Centimeter-Scale Distance

The transmon, a fabrication-friendly superconducting qubit, remains a leading candidate for scalable quantum computing. Recent advances in tunable couplers have accelerated progress toward high-performance quantum processors. However, extending coherent interactions beyond millimeter scales to enhance quantum connectivity presents a critical challenge. Here, we introduce a hybrid-mode coupler exploiting resonator-transmon hybridization to simultaneously engineer the two lowest-frequency mode, enabling high-contrast coupling between centimeter-scale transmons. For a 1-cm coupler, our framework predicts flux-tunable $XX$ and $ZZ$ coupling strengths reaching 23 MHz and 100 MHz, with modulation contrasts exceeding $10^2$ and $10^4$, respectively, demonstrating quantitative agreement with an effective two-channel model. This work provides an efficient pathway to mitigate the inherent connectivity constraints imposed by short-range interactions, enabling transmon-based architectures compatible with hardware-efficient quantum tasks.

quant-ph↗

arXiv2025

Saturating the Quantum Cramér--Rao Bound in Prioritised Parameter Estimation

Measurement incompatibility is a cornerstone of quantum mechanics. In the context of estimating multiple parameters of a quantum system, this manifests as a fundamental trade-off between the precisions with which different parameters can be estimated. Often, a parameter can be optimally measured, but at the cost of gaining no information about incompatible parameters. Here, we report that there are systems where one parameter's information can be maximised while not completely losing information about the other parameters. In doing so, we find attainable trade-off relations for quantum parameter estimation with a structure that is different to typical Heisenberg-type trade-offs. We demonstrate our findings by implementing an optimal entangling measurement on a Quantinuum trapped-ion quantum computer.

quant-ph↗

arXiv2025

Path-Consistency with Prefix Enhancement for Efficient Inference in LLMs

To enhance the reasoning capabilities of large language models (LLMs), self-consistency has become a popular approach, combining multiple samplings with majority voting. However, current methods are computationally expensive and time-consuming due to the need for numerous samplings. To address this, this paper introduces path-consistency, which leverages the confidence of earlier-generated answers to identify the most promising prefix and guide the generation of subsequent branches. By dynamically guiding the generation of subsequent branches based on this prefix, path-consistency mitigates both the errors and redundancies from random or less useful sampling in self-consistency. This approach reduces errors and redundancies from random sampling, significantly accelerating inference by minimizing token consumption. Our extensive empirical results demonstrate that path-consistency improves inference latency by up to 40.5\%, while maintaining task accuracy across various tasks, including mathematical reasoning, commonsense reasoning, and symbolic reasoning.

cs.CL↗

arXiv2025

Generalized collective quantum tomography: algorithm design, optimization, and validation

Quantum tomography is a fundamental technique for characterizing, benchmarking, and verifying quantum states and devices. It plays a crucial role in advancing quantum technologies and deepening our understanding of quantum mechanics. Collective quantum state tomography, which estimates an unknown state \r{ho} through joint measurements on multiple copies $ρ\otimes\cdots\otimesρ$ of the unknown state, offers superior information extraction efficiency. Here we extend this framework to a generalized setting where the target becomes $S_1\otimes\cdots\otimes S_n$, with each $S_i$ representing identical or distinct quantum states, detectors, or processes from the same category. We formulate these tasks as optimization problems and develop three algorithms for collective quantum state, detector and process tomography, respectively, each accompanied by an analytical characterization of the computational complexity and mean squared error (MSE) scaling. Furthermore, we develop optimal solutions of these optimization problems using sum of squares (SOS) techniques with semi-algebraic constraints. The effectiveness of our proposed methods is demonstrated through numerical examples. Additionally, we experimentally demonstrate the algorithms using two-copy collective measurements, where entangled measurements directly provide information about the state purity. Compared to existing methods, our algorithms achieve lower MSEs and approach the collective MSE bound by effectively leveraging purity information.

quant-ph↗

arXiv2025

TrajAgent: An LLM-Agent Framework for Trajectory Modeling via Large-and-Small Model Collaboration

Trajectory modeling, which includes research on trajectory data pattern mining and future prediction, has widespread applications in areas such as life services, urban transportation, and public administration. Numerous methods have been proposed to address specific problems within trajectory modeling. However, the heterogeneity of data and the diversity of trajectory tasks make effective and reliable trajectory modeling an important yet highly challenging endeavor, even for domain experts. In this paper, we propose TrajAgent, an agent framework powered by large language models, designed to facilitate robust and efficient trajectory modeling through automation modeling. This framework leverages and optimizes diverse specialized models to address various trajectory modeling tasks across different datasets effectively. In TrajAgent, we first develop UniEnv, an execution environment with a unified data and model interface, to support the execution and training of various models. Building on UniEnv, we introduce an agentic workflow designed for automatic trajectory modeling across various trajectory tasks and data. Furthermore, we introduce collaborative learning schema between LLM-based agents and small speciallized models, to enhance the performance of the whole framework effectively. Extensive experiments on five tasks using four real-world datasets demonstrate the effectiveness of TrajAgent in automated trajectory modeling, achieving a performance improvement of 2.38%-69.91% over baseline methods. The codes and data can be accessed via https://github.com/tsinghua-fib-lab/TrajAgent.

cs.CL↗

arXiv2025

All-Gaussian State Discrimination Beyond the Coherent Helstrom Bound

A core problem in communications is the optimal discrimination of binary-phase-shift-keyed (BPSK) signals. A longstanding goal has been to reach the fundamental quantum limit, known as the Helstrom bound, for BPSK signals encoded in coherent states. However, due to technical constraints, proposals for reaching the bound remain impractical. In this letter we take an alternative approach: using only Gaussian optics - displaced squeezed states and homodyne detection - we achieve discrimination of BPSK signals with error rates below what can be achieved using coherent states and any quantum measurement.

quant-ph↗

arXiv2025

Structure-Aware Cooperative Ensemble Evolutionary Optimization on Combinatorial Problems with Multimodal Large Language Models

Evolutionary algorithms (EAs) have proven effective in exploring the vast solution spaces typical of graph-structured combinatorial problems. However, traditional encoding schemes, such as binary or numerical representations, often fail to straightforwardly capture the intricate structural properties of networks. Through employing the image-based encoding to preserve topological context, this study utilizes multimodal large language models (MLLMs) as evolutionary operators to facilitate structure-aware optimization over graph data. To address the visual clutter inherent in large-scale network visualizations, we leverage graph sparsification techniques to simplify structures while maintaining essential structural features. To further improve robustness and mitigate bias from different sparsification views, we propose a cooperative evolutionary optimization framework that facilitates cross-domain knowledge transfer and unifies multiple sparsified variants of diverse structures. Additionally, recognizing the sensitivity of MLLMs to network layout, we introduce an ensemble strategy that aggregates outputs from various layout configurations through consensus voting. Finally, experiments on real-world networks through various tasks demonstrate that our approach improves both the quality and reliability of solutions in MLLM-driven evolutionary optimization.

cs.NE↗