arXiv subjects

Junjie Liu

Publications and source records attributed to Junjie Liu.

At least 37 records · Page 2Linked to original sources

arXiv2026

The Second Challenge on Cross-Domain Few-Shot Object Detection at NTIRE 2026: Methods and Results

Cross-domain few-shot object detection (CD-FSOD) remains a challenging problem for existing object detectors and few-shot learning approaches, particularly when generalizing across distinct domains. As part of NTIRE 2026, we hosted the second CD-FSOD Challenge to systematically evaluate and promote progress in detecting objects in unseen target domains under limited annotation conditions. The challenge received strong community interest, with 128 registered participants and a total of 696 submissions. Among them, 31 teams actively participated, and 19 teams submitted valid final results. Participants explored a wide range of strategies, introducing innovative methods that push the performance frontier under both open-source and closed-source tracks. This report presents a detailed overview of the NTIRE 2026 CD-FSOD Challenge, including a summary of the submitted approaches and an analysis of the final results across all participating teams. Challenge Codes: https://github.com/ohMargin/NTIRE2026_CDFSOD.

cs.CV↗

arXiv2026

Direct temperature readout in nonequilibrium quantum thermometry

Quantum thermometry aims to measure temperature in nanoscale quantum systems, paralleling classical thermometry. However, temperature is not a quantum observable, and most theoretical studies have therefore concentrated on analyzing fundamental precision limits set by the quantum Fisher information through the quantum Cramer-Rao bound. In contrast, whether a direct temperature readout can be achieved in quantum thermometry remains largely unexplored, particularly under the nonequilibrium conditions prevalent in real-world applications. To address this, we develop a direct temperature readout scheme based on a thermodynamic inference strategy. The scheme integrates two conceptual developments: (i) By applying the maximum entropy principle with the thermometer's mean energy as a constraint, we assign a reference temperature to the nonequilibrium thermometer. We demonstrate that this reference temperature outperforms a commonly used effective temperature defined through equilibrium analogy. (ii) We obtain positive semi-definite error functions that lower-bound the deviation of the reference temperature from the true temperature-in analogy to the quantum Cramer-Rao bound for the mean squared error-and vanish upon thermalization with the sample. Combining the reference temperature with these error functions, we introduce a notion of corrected dynamical temperature which furnishes a postprocessed temperature readout under nonequilibrium conditions. This corrected dynamical temperature can be evaluated adaptively without prior knowledge of the actual temperature. We validate the corrected dynamical temperature in a qubit-based thermometer under a range of nonequilibrium initial states, confirming its capability to estimate the true temperature. Importantly, we find that increasing quantum coherence can enhance the precision of this readout.

quant-ph↗

arXiv2026

Dynamic Whole-Body Dancing with Humanoid Robots -- A Model-Based Control Approach

This paper presents an integrated model-based framework for generating and executing dynamic whole-body dance motions on humanoid robots. The framework operates in two stages: offline motion generation and online motion execution, both leveraging future state prediction to enable robust and dynamic dance motions in real-world environments. In the offline motion generation stage, human dance demonstrations are captured via a motion capture (MoCap) system, retargeted to the robot by solving a Quadratic Programming (QP) problem, and further refined using Trajectory Optimization (TO) to ensure dynamic feasibility. In the online motion execution stage, a centroidal dynamics-based Model Predictive Control (MPC) framework tracks the planned motions in real time and proactively adjusts swing foot placement to adapt to real world disturbances. We validate our framework on the full-size humanoid robot Kuavo 4Pro, demonstrating the dynamic dance motions both in simulation and in a four-minute live public performance with a team of four robots. Experimental results show that longer prediction horizons improve both motion expressiveness in planning and stability in execution.

cs.RO↗

arXiv2026

Efficient RGB-D Scene Understanding via Multi-task Adaptive Learning and Cross-dimensional Feature Guidance

Scene understanding plays a critical role in enabling intelligence and autonomy in robotic systems. Traditional approaches often face challenges, including occlusions, ambiguous boundaries, and the inability to adapt attention based on task-specific requirements and sample variations. To address these limitations, this paper presents an efficient RGB-D scene understanding model that performs a range of tasks, including semantic segmentation, instance segmentation, orientation estimation, panoptic segmentation, and scene classification. The proposed model incorporates an enhanced fusion encoder, which effectively leverages redundant information from both RGB and depth inputs. For semantic segmentation, we introduce normalized focus channel layers and a context feature interaction layer, designed to mitigate issues such as shallow feature misguidance and insufficient local-global feature representation. The instance segmentation task benefits from a non-bottleneck 1D structure, which achieves superior contour representation with fewer parameters. Additionally, we propose a multi-task adaptive loss function that dynamically adjusts the learning strategy for different tasks based on scene variations. Extensive experiments on the NYUv2, SUN RGB-D, and Cityscapes datasets demonstrate that our approach outperforms existing methods in both segmentation accuracy and processing speed.

cs.CV↗

arXiv2026

Seeking Necessary and Sufficient Information from Multimodal Medical Data

Learning multimodal representations from medical images and other data sources can provide richer information for decision-making. While various multimodal models have been developed for this, they overlook learning features that are both necessary (must be present for the outcome to occur) and sufficient (enough to determine the outcome). We argue learning such features is crucial as they can improve model performance by capturing essential predictive information, and enhance model robustness to missing modalities as each modality can provide adequate predictive signals. Such features can be learned by leveraging the Probability of Necessity and Sufficiency (PNS) as a learning objective, an approach that has proven effective in unimodal settings. However, extending PNS to multimodal scenarios remains underexplored and is non-trivial as key conditions of PNS estimation are violated. We address this by decomposing multimodal representations into modality-invariant and modality-specific components, then deriving tractable PNS objectives for each. Experiments on synthetic and real-world medical datasets demonstrate our method's effectiveness. Code will be available on GitHub.

cs.CV↗

arXiv2026

LIBERO-X: Robustness Litmus for Vision-Language-Action Models

Reliable benchmarking is critical for advancing Vision-Language-Action (VLA) models, as it reveals their generalization, robustness, and alignment of perception with language-driven manipulation tasks. However, existing benchmarks often provide limited or misleading assessments due to insufficient evaluation protocols that inadequately capture real-world distribution shifts. This work systematically rethinks VLA benchmarking from both evaluation and data perspectives, introducing LIBERO-X, a benchmark featuring: 1) A hierarchical evaluation protocol with progressive difficulty levels targeting three core capabilities: spatial generalization, object recognition, and task instruction understanding. This design enables fine-grained analysis of performance degradation under increasing environmental and task complexity; 2) A high-diversity training dataset collected via human teleoperation, where each scene supports multiple fine-grained manipulation objectives to bridge the train-evaluation distribution gap. Experiments with representative VLA models reveal significant performance drops under cumulative perturbations, exposing persistent limitations in scene comprehension and instruction grounding. By integrating hierarchical evaluation with diverse training data, LIBERO-X offers a more reliable foundation for assessing and advancing VLA development.

cs.CV↗

arXiv2026

Quantum information-cost relations and fluctuations beyond thermal environments: A thermodynamic inference approach

The Landauer's principle, a cornerstone of information thermodynamics, provides a fundamental lower bound on the energetic cost of information erasure in terms of the information content change. However, its traditional formulation is largely confined to systems exchanging solely energy with an ideal thermal bath. In this work, we derive general information-cost trade-off relations that go beyond the scope of Landauer's principle by developing a thermodynamic inference approach based on the maximum entropy principle. These relations require only information about the system and are applicable to complex quantum scenarios involving multiple conserved charges and non-thermal environments. Specifically, we present two key results: (i) In scenarios where only the mean values of observables are accessible, we derive an information-content-informed upper bound on the thermodynamic cost which complements an existing generalized Landauer lower bound. (ii) When second-order fluctuations can also be measured, we obtain an information-content-informed lower bound on the change in variances of observables, thereby extending the Landauer's principle to constrain higher-order fluctuation costs. We numerically validate our information-cost trade-off relations using a coupled-qubit system exchanging energy and excitations, a driven qubit implementing an information erasure process, and a driven double quantum dot system that can operate as an inelastic heat engine. Our results underscore the broad utility of maximum-entropy inference in constraining thermodynamic costs for generic finite-time quantum processes, with direct relevance to quantum information processing and quantum thermodynamic applications.

quant-ph↗

arXiv2026

Quantum Interaction Between Free Electrons and Light Involving First-order and Second-order Process

Photon-induced Near-field Electron Microscopy (PINEM) effect has revealed the quantum interaction between free electrons and optical near filed, which demonstrated plenty of novel phenomena of manipulating free electron wave packet and detecting/shaping quantum photonic states. However, free electrons generally only absorb/emit one photon at a time, while the physical mechanism and phenomena of free electron-two-photon interaction have not been studied yet. Moreover, the relationship between PINEM and Kapitza-Dirac (KD) effect and nonlinear Compton scattering is still unclear. Here we develop the full quantum theory of electron-photon interaction considering the two-photon process. It is revealed that the emission/absorption of two photons by electrons can be greatly enhanced by manipulating the electric field component of optical near field, and the quantum interference between single-photon and two-photon processes can occur in some circumstances, which affects the photon number state, electron energy states and electron-photon entanglement. Meanwhile, it is found that the KD effect (elastic electron-photon scattering) and nonlinear Compton scattering (inelastic electron-photon scattering) are also a kind of two-photon process and the distribution of electrons can be deduced analytically based on the full quantum theory. Our work uncovers the possible abundant phenomena when free electron interacting with two photons, paves the way for more in-depth studies of nonlinear processes in electron-photon quantum interactions in the future.

quant-ph↗

arXiv2025

TCFormer: A 5M-Parameter Transformer with Density-Guided Aggregation for Weakly-Supervised Crowd Counting

Crowd counting typically relies on labor-intensive point-level annotations and computationally intensive backbones, restricting its scalability and deployment in resource-constrained environments. To address these challenges, this paper proposes the TCFormer, a tiny, ultra-lightweight, weakly-supervised transformer-based crowd counting framework with only 5 million parameters that achieves competitive performance. Firstly, a powerful yet efficient vision transformer is adopted as the feature extractor, the global context-aware capabilities of which provides semantic meaningful crowd features with a minimal memory footprint. Secondly, to compensate for the lack of spatial supervision, we design a feature aggregation mechanism termed the Learnable Density-Weighted Averaging module. This module dynamically re-weights local tokens according to predicted density scores, enabling the network to adaptively modulate regional features based on their specific density characteristics without the need for additional annotations. Furthermore, this paper introduces a density-level classification loss, which discretizes crowd density into distinct grades, thereby regularizing the training process and enhancing the model's classification power across varying levels of crowd density. Therefore, although TCformer is trained under a weakly-supervised paradigm utilizing only image-level global counts, the joint optimization of count and density-level losses enables the framework to achieve high estimation accuracy. Extensive experiments on four benchmarks including ShanghaiTech A/B, UCF-QNRF, and NWPU datasets demonstrate that our approach strikes a superior trade-off between parameter efficiency and counting accuracy and can be a good solution for crowd counting tasks in edge devices.

cs.CV↗

arXiv2025

Frustratingly Easy Task-aware Pruning for Large Language Models

Pruning provides a practical solution to reduce the resources required to run large language models (LLMs) to benefit from their effective capabilities as well as control their cost for training and inference. Research on LLM pruning often ranks the importance of LLM parameters using their magnitudes and calibration-data activations and removes (or masks) the less important ones, accordingly reducing LLMs' size. However, these approaches primarily focus on preserving the LLM's ability to generate fluent sentences, while neglecting performance on specific domains and tasks. In this paper, we propose a simple yet effective pruning approach for LLMs that preserves task-specific capabilities while shrinking their parameter space. We first analyze how conventional pruning minimizes loss perturbation under general-domain calibration and extend this formulation by incorporating task-specific feature distributions into the importance computation of existing pruning algorithms. Thus, our framework computes separate importance scores using both general and task-specific calibration data, partitions parameters into shared and exclusive groups based on activation-norm differences, and then fuses their scores to guide the pruning process. This design enables our method to integrate seamlessly with various foundation pruning techniques and preserve the LLM's specialized abilities under compression. Experiments on widely used benchmarks demonstrate that our approach is effective and consistently outperforms the baselines with identical pruning ratios and different settings.

cs.CL↗

arXiv2025

Electric-field Quantum Sensing Exploiting a Photogenerated Charge-transfer Triplet State in a Molecular Semiconductor

Molecular spin systems are promising platforms for quantum sensing due to their chemically tunable Hamiltonians, enabling tailored coherence properties and interactions with external fields. However, electric field sensing remains challenging owing to typically weak spin-electric coupling (SEC) and limited directional sensitivity. Addressing these issues using heavy atoms exhibiting strong atomic spin-orbit couplings (SOC) often compromises spin coherence times. Here, we demonstrate coherent electric field sensing using a photogenerated charge-transfer (CT) spin triplet state in the organic molecule ACRSA (10-phenyl-10H,10' H-spiro\[acridine-9,9'-anthracen]-10'-one). By embedding electric field pulses within a Hahn echo sequence, we coherently manipulate the spin triplet and extract both the magnitude and directional dependence of its SEC. The measured SEC strength is approximately 0.51 Hz/(V/m), comparable to values reported in systems with strong atomic SOC, illustrating that heavy atoms are not a prerequisite for electric-field sensitivity of spin states. Our findings position organic CT triplets as chemically versatile and directionally sensitive quantum sensors of E-fields that function without atomic-SOC-mediated mechanisms.

quant-ph↗

arXiv2025

Tracing Partisan Bias to Its Emotional Fingerprints: A Computational Approach to Mitigation

This study introduces a novel framework for analysing and mitigating media bias by tracing partisan stances to their linguistic roots in emotional language. We posit that partisan bias is not merely an abstract stance but materialises as quantifiable 'emotional fingerprints' within news texts. These fingerprints are systematically measured using the Valence-Arousal-Dominance (VAD) framework, allowing us to decode the affective strategies behind partisan framing. Our analysis of the Allsides dataset confirms this hypothesis, revealing distinct and statistically significant emotional fingerprints for left, centre, and right-leaning media. Based on this evidence-driven approach, we then propose a computational approach to mitigation through NeutraSum, a model designed to neutralise these identified emotional patterns. By explicitly targeting the VAD characteristics of biased language, NeutraSum generates summaries that are not only coherent but also demonstrably closer to an emotionally neutral baseline. Experimental results validate our framework: NeutraSum successfully erases the partisan emotional fingerprints from its summaries, achieving a demonstrably lower emotional bias score than other models. This work pioneers a new path for bias mitigation, shifting the focus from treating symptoms (political labels) to addressing the cause: the emotional encoding of partisan bias in language.

cs.CL↗

arXiv2025

Quantum Mpemba effect in parity-time symmetric systems

The quantum Mpemba effect (QMPE), an anomalous relaxation phenomenon, has been demonstrated in both closed and open Hermitian quantum systems. While some studies have linked the QMPE to Liouvillian exceptional points--non-Hermitian features emerged at the Liouvillian level--in open Hermitian quantum systems, it remains largely unexplored whether the QMPE can occur in intrinsic non-Hermitian systems, where non-Hermiticity is inherent at the Hamiltonian level. Here, we demonstrate unequivocally the occurrence of QMPE in experimentally realizable parity-time-symmetric qubit systems immersed in a bosonic bath. Using established quantifiers for QMPE, we show numerically that the QMPE persists across parameter regimes both near and far from Hamiltonian and Liouvillian exceptional points, but disappears entirely when Hermitian Hamiltonian is restored. Interestingly, neither Hamiltonian nor Liouvillian exceptional points demarcate the boundaries of the QMPE regime. To complement numerical results, we develop an analytical description based on a long-time approximation of the relaxation dynamics of quantifiers. This approach allows us to decipher the number of intersections between two dynamical trajectories of quantifier starting from two initial conditions in the validity regime of the long-time approximation, thereby providing additional information that delineate the parameter regimes supporting the genuine QMPE. We further demonstrate the robustness of QMPE against increasing the number of qubits and dephasing effect. Our findings not only broaden the scope of the QMPE but also suggest its intricate interplay with non-Hermitian features beyond exceptional points.

quant-ph↗

arXiv2025

GeoCAD: Local Geometry-Controllable CAD Generation with Large Language Models

Local geometry-controllable computer-aided design (CAD) generation aims to modify local parts of CAD models automatically, enhancing design efficiency. It also ensures that the shapes of newly generated local parts follow user-specific geometric instructions (e.g., an isosceles right triangle or a rectangle with one corner cut off). However, existing methods encounter challenges in achieving this goal. Specifically, they either lack the ability to follow textual instructions or are unable to focus on the local parts. To address this limitation, we introduce GeoCAD, a user-friendly and local geometry-controllable CAD generation method. Specifically, we first propose a complementary captioning strategy to generate geometric instructions for local parts. This strategy involves vertex-based and VLLM-based captioning for systematically annotating simple and complex parts, respectively. In this way, we caption $\sim$221k different local parts in total. In the training stage, given a CAD model, we randomly mask a local part. Then, using its geometric instruction and the remaining parts as input, we prompt large language models (LLMs) to predict the masked part. During inference, users can specify any local part for modification while adhering to a variety of predefined geometric instructions. Extensive experiments demonstrate the effectiveness of GeoCAD in generation quality, validity and text-to-CAD consistency. Code will be available at https://github.com/Zhanwei-Z/GeoCAD.

cs.CV↗

arXiv2025

Full counting statistics of electron-photon hybrid systems: Joint statistics and fluctuation symmetry

Electron-photon hybrid systems serve as ideal light-matter interfaces with broad applications in quantum technologies. These systems are typically operated dynamically under nonequilibrium conditions, giving rise to coupled electronic and photonic currents. Understanding the joint fluctuation behavior of these currents is essential for assessing the performance of light-matter interfaces that rely on electron-photon correlations. Here, we investigate the full counting statistics of coupled electronic and photonic currents in an experimentally feasible hybrid system composed of a double quantum dot coupled to an optical cavity. We employ the framework of quantum Lindblad master equation which is augmented with both electronic and photonic counting fields to derive their joint cumulant generating function--a treatment that differs significantly from existing studies, which typically focus on either electron or photon statistics separately. We reveal that the ratio between photonic and electronic currents, as well as their variances, can deviate from an expected quadratic scaling law in the large electron-photon coupling regime. Furthermore, we demonstrate that conventional modelings of photonic dissipation channels in quantum master equations must be modified to ensure that the joint cumulant generating function satisfies the fluctuation symmetry enforced by the fluctuation theorem. Our results advance the understanding of joint fluctuation behaviors in electron-photon hybrid systems and may inform the design of efficient quantum light-matter interfaces.

cond-mat.stat-mech↗

arXiv2025

Programmable self-assembly for scalable molecular electronic architectures

Molecular electronics and other technologies whose components comprise individual molecules have been pursued for half a century because the molecular scale represents the limit of miniaturisation of objects whose structure is tuneable for function. Despite the promise, practical progress has been hindered by the lack of methodologies for directed assembly of arbitrary structures applicable at the molecular scale. DNA nanotechnology is an emerging framework that uses programmed synthetic oligomers to encode the design of self-assembling structures with atomic precision at the nanoscale. Here, we leverage DNA-directed self-assembly to construct single-molecule electrical transport devices in high yield, precisely positioning a metal-porphyrin between two 60 nm gold nanoparticles. Following deposition on SiO2 substrates, we image and establish electrical contact via established nanofabrication techniques. Each step of the process has a high success rate and we demonstrate device yields dramatically better than is possible using conventional approaches. Our approach is inherently scalable and adaptable to devices incorporating multiple heterogenous functional molecular components, finally offering a realistic framework for the realisation of classical and quantum molecular technologies.

cond-mat.mes-hall↗

arXiv2025

An Error Model for Evaluating the Accuracy of Satellite-Based XCO$_2$ Products

Several satellites (e.g., OCO-2 & 3) and their derived products now provide spatially extensive coverage of the abundance of carbon dioxide in the atmospheric column (XCO$_2$). However, the accuracy of the XCO$_2$ reported in these products needs to be carefully assessed for any downstream scientific analysis; this involves comparison with reference datasets, such as those from the Total Carbon Column Observing Network (TCCON). Previously, systematic and random errors have been used to quantify differences between satellite-based XCO$_2$ measurements and TCCON data. The spatiotemporal density of satellite observations enables the decomposition of the error variability into these components. This study aims to unify the definitions of these error components through a hierarchical statistical model with explicit mathematical terms, which enables a formal definition of the underlying assumptions and estimation of each component. Specifically, we focus on defining model elements, like global bias and systematic and random error, as part of this framework. We use it to compare OCO-2 XCO$_2$ v11.1 data (both original scenes from the `Lite' files and 10-sec averages) and gridded Making Earth System Data Records for Use in Research Environments (MEaSUREs) products to TCCON data. The MEaSUREs products exhibit comparable systematic errors to other OCO-2 products, with larger errors over land versus ocean. We describe the methodology for creating the MEaSUREs products, including their prior and posterior error covariances, with information on spatial correlation for efficient incorporation into scientific analysis.

stat.AP↗

arXiv2025

Anomalous flow in correlated quantum systems: No-go result and multiple-charge scenario

Correlated quantum systems can exhibit thermodynamic behaviors that defy classical expectations, with anomalous energy flow (AEF) against temperature gradients serving as a paradigmatic example. While AEF has been shown to arise from the consumption of initial quantum correlations, little is known about whether AEF can occur without correlation depletion, or if analogous anomalous transport exists for conserved quantities--dubbed charges--other than energy. Here, we develop a general global-local thermodynamic approach to describe charge exchange between arbitrary correlated quantum systems. For energy-conserving systems, we analytically rule out AEF in initially uncorrelated states, even with the involvement of quantum catalysts, thereby complementing existing studies. In contrast, in systems with multiple conserved charges, we uncover a mechanism for AEF that requires no initial correlations but is instead induced by a drag effect from normal flows of non-energy charges. Furthermore, by treating all conserved charges on equal footing, we generalize AEF to a broader concept of anomalous charge flow, applicable to any conserved charge. We confirm theoretical expectations with numerical examples. These findings deepen our understanding of nonequilibrium quantum thermodynamics and open new avenues for controlling transport phenomena in correlated quantum systems.

quant-ph↗