arXiv subjects

Zhihong Liu

Publications and source records attributed to Zhihong Liu.

At least 19 recordsLinked to original sources

arXiv2026

Dilaton gravity can enhance quantum coherence and reduce entanglement

We investigate the influence of the Garfinkle-Horowitz-Strominger (GHS) dilaton black hole on different quantum resources of Dirac fields beyond the single-mode approximation. By employing the negativity to characterize quantum entanglement and the $l_1$-norm and the relative entropy of coherence to characterize quantum coherence, we demonstrate that these resources exhibit remarkably different responses to the gravitational field. Specifically, increasing the dilaton parameter continuously suppresses quantum entanglement, leaving only a finite residual amount in the strong-gravity regime, whereas quantum coherence is enhanced, indicating that the dilaton-induced spacetime affects nonlocal quantum correlations and local quantum superposition in fundamentally different ways. Furthermore, we show that an initially maximally entangled state does not always possess the largest negativity after propagating in the GHS dilaton spacetime; instead, under appropriate conditions, certain non-maximally entangled states can retain stronger entanglement than the maximally entangled one. These findings reveal the resource-dependent nature of gravitational effects in dilaton black hole backgrounds and provide new insights into the manipulation and protection of quantum resources for relativistic quantum information processing in curved spacetime.

gr-qc↗

arXiv2026

Is Retrieval All You Need? Assessment and Emergence of Novelty in Protein Structure Generation

Protein backbone generation models are often credited with exploring novel fold space based solely on low full-chain similarity to known proteins, yet this cannot distinguish a genuinely new fold from a novel assembly of known structural units. We first ask whether this granularity mismatch alone explains the reported rates, and introduce the Domain Retrieval Rate (DRR), the fraction of generated backbones for which any constituent domain matches a known domain in CATH S40. Applied to eight backbone generation models spanning diffusion and flow-matching paradigms, DRR finds locally alignable known structure in most outputs, while the fraction containing a substantially covered complete domain is considerably smaller and depends on the scoring convention. To calibrate what retrieval alone can achieve, we propose RetFold, a zero-training baseline that constructs backbones by retrieving CATH domains and refining inter-domain connections through geometry-based helix-linker optimization, at two orders of magnitude lower cost on CPU alone.

q-bio.BM↗

arXiv2026

Any House Any Task: Scalable Long-Horizon Planning for Abstract Human Tasks

Open world language conditioned task planning is crucial for robots operating in large-scale household environments. While many recent works attempt to address this problem using Large Language Models (LLMs) via prompting or training, a key challenge remains scalability. Performance often degrades rapidly with increasing environment size, plan length, instruction ambiguity, and constraint complexity. In this work, we propose Any House Any Task (AHAT), a household task planner optimized for long-horizon planning in large environments given ambiguous human instructions. At its core, AHAT utilizes an LLM trained to map task instructions and textual scene graphs into grounded subgoals defined in the Planning Domain Definition Language (PDDL). These subgoals are subsequently solved to generate feasible and optimal long-horizon plans through explicit symbolic reasoning. To enhance the model's ability to decompose complex and ambiguous intentions, we introduce TGPO, a novel reinforcement learning algorithm that integrates external correction of intermediate reasoning traces into Group Relative Policy Optimization (GRPO). Experiments demonstrate that AHAT achieves significant performance gains over state-of-the-art prompting, planning, and learning methods, particularly in human-style household tasks characterized by brief instructions but requiring complex execution plans.

cs.RO↗

arXiv2026

MirrorPPR: Exemplar-Based Portrait Photo Retouching

While text-guided image editing has made remarkable progress, it remains limited in structural portrait retouching. Textual descriptions struggle to convey fine-grained changes to facial features and body proportions. To address this gap, we introduce Exemplar-Based Portrait Photo Retouching, where the model is given an exemplar pair and tasked with inferring and applying the same retouching operations to a new query image. Existing exemplar-based editing methods primarily focus on tasks with pronounced visual transformations. In contrast, structural portrait retouching involves extremely delicate and localized modifications, making accurate extraction and transfer of these edits challenging. To tackle this, we propose MirrorPPR, a novel framework designed to capture and transfer subtle structural retouching operations. Our method uses a Retouching Operation Extractor to capture the subtle differences from the exemplar pair. The extracted representations are then injected into a pre-trained Diffusion Transformer (DiT) through a connector and Low-Rank Adaptation (LoRA) modules. Furthermore, constructing perfectly aligned cross-identity training pairs is severely hindered by operation misalignment. To overcome this, we propose an advanced data self-augmentation paradigm that ensures strictly aligned retouching operations. To alleviate data scarcity and support this novel task, we introduce MirrorPPR47M, a large-scale dataset with over 47 million retouched pairs. By structuring the dataset into simulated and professional subsets, we enable progressive curriculum learning to smoothly optimize the network. Extensive experiments demonstrate that MirrorPPR significantly outperforms existing baselines in both retouching quality and identity preservation. The project page is available at https://sjtu-deng-lab.github.io/MirrorPPR.

cs.CV↗

arXiv2026

Reflecting boundary induced modulation of tripartite coherence harvesting

We study the extraction of quantum coherence by three static Unruh-DeWitt (UDW) detectors that interact locally with a massless scalar vacuum field in the vicinity of an infinite perfectly reflecting boundary. Depending on the setup, the detectors are positioned either parallel or orthogonal to the boundary, with their energy gaps chosen to satisfy the hierarchy $Ω_C\geq Ω_B\geq Ω_A$. Our analysis reveals that decreasing the detector-boundary separation leads to a monotonic degradation of quantum coherence, whereas the same boundary effect can simultaneously preserve and even amplify the harvested quantum entanglement. Moreover, when the detectors possess distinct energy gaps, coherence extraction is further inhibited; strikingly, such non-identical configurations substantially enhance the efficiency of entanglement harvesting and markedly extend the range of detector separations over which non-negligible entanglement can be generated. Nevertheless, the harvesting of nonlocal quantum coherence is achievable over a significantly broader range of detector separations than that of quantum entanglement. Despite exhibiting similar overall behavior, orthogonal detector configurations outperform parallel ones in coherence harvesting, highlighting the quantitative influence of detector geometry. Overall, our study reveals a hierarchical distinction between quantum coherence and entanglement as operational resources in structured vacuum fields: quantum coherence is not only more readily accessible across space but also more robust than entanglement, whereas entanglement exhibits richer features and can be selectively activated and enhanced through boundary effects and detector non-uniformity.

quant-ph↗

arXiv2026

World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis

We propose world-language-action (WLA) models as a new class of embodied foundation models. WLA takes textual instructions, images, and robot states as inputs to jointly predict textual subtasks, subgoal images, and robot actions, conjoining the \emph{world modeling interface} to learn from extensive egocentric videos as in the world-action model (WAM) and the \emph{language reasoning} capacities to solve complex long-horizon tasks as in vision-language-action (VLA) models. At the core of WLA lies an \emph{autoregressive (AR)} Transformer backbone, instead of a bidirectional diffusion Transformer as in WAMs, to predict the \emph{next state}, comprising the \emph{semantic-level} textual intention and complementary \emph{fine-grained} physical dynamics. The physical dynamics are supervised by the world modeling objective based on a dedicated World Expert, and are leveraged to ease the characterization of the state-action correlation for the Action Expert. WLA leverages meta-queries to make the world prediction \emph{implicitly} impact the action generation so that the former can be disabled during inference. The world prediction can also be activated to enable test-time scaling for improved robot control. Our WLA-0 prototype, with 2B active parameters, achieves 40 ms per inference on an NVIDIA RTX 5090. Evaluations across simulated and real-world environments demonstrate that WLA-0 achieves state-of-the-art multi-task and long-horizon learning abilities, e.g., 92.94\% success rate on RoboTwin2.0 Clean and 56.5\% success rate on RMBench. WLA-0 also holds the promise to learn novel tasks directly from \emph{cross-embodiment robot videos} without action annotations.

cs.RO↗

arXiv2026

ProductWebGen: Benchmarking Multimodal Product Webpage Generation

Crafting a product display webpage from a source product image, along with layout and visual content instructions, holds significant practical value for domains such as marketing, advertising, and E-commerce. Intuitively, this task demands strict visual consistency across product displays and high-fidelity instruction following to jointly generate renderable HTML code. These requirements on controllability and instruction-following are closely aligned with the core features of advanced multimodal generative models, such as image editing models and unified models. To this end, this paper introduces ProductWebGen to systematically benchmark the product webpage generation capacities of these models. We organize ProductWebGen with 500 test samples covering 13 product categories; each sample consists of a source image, a visual content instruction, and a webpage instruction. The task is to generate a product showcase webpage including multiple consistent images in accordance with the source image and instructions. Given the mixed-modality input-output nature of the task, we design and systematically compare two workflows for evaluation -- one uses large language models and image editing models to separately generate HTML code and images (editing-based), while the other relies on a single UM to generate both, with image generation conditioned on the preceding multimodal context (UM-based). Empirical results show that editing-based approaches achieve leading results in webpage instruction following and content appeal, while UM-based ones may display more advantages in fulfilling visual content instructions. We also construct a supervised fine-tuning dataset, ProductWebGen-1k, with 1,000 groups of real product images and LLM-generated HTML code. We verify its effectiveness on the open-source UM BAGEL. The data and code are available at https://github.com/SJTU-DENG-Lab/ProductWebGen.

cs.CV↗

arXiv2026

Task-Guided Prompting for Unified Remote Sensing Image Restoration

Remote sensing image restoration (RSIR) is essential for recovering high-fidelity imagery from degraded observations, enabling accurate downstream analysis. However, most existing methods focus on single degradation types within homogeneous data, restricting their practicality in real-world scenarios where multiple degradations often across diverse spectral bands or sensor modalities, creating a significant operational bottleneck. To address this fundamental gap, we propose TGPNet, a unified framework capable of handling denoising, cloud removal, shadow removal, deblurring, and SAR despeckling within a single, unified architecture. The core of our framework is a novel Task-Guided Prompting (TGP) strategy. TGP leverages learnable, task-specific embeddings to generate degradation-aware cues, which then hierarchically modulate features throughout the decoder. This task-adaptive mechanism allows the network to precisely tailor its restoration process for distinct degradation patterns while maintaining a single set of shared weights. To validate our framework, we construct a unified RSIR benchmark covering RGB, multispectral, SAR, and thermal infrared modalities for five aforementioned restoration tasks. Experimental results demonstrate that TGPNet achieves state-of-the-art performance on both unified multi-task scenarios and unseen composite degradations, surpassing even specialized models in individual domains such as cloud removal. By successfully unifying heterogeneous degradation removal within a single adaptive framework, this work presents a significant advancement for multi-task RSIR, offering a practical and scalable solution for operational pipelines. The code and benchmark will be released at https://github.com/huangwenwenlili/TGPNet.

eess.IV↗

arXiv2026

Online Statistical Inference for Contextual Bandits via Stochastic Gradient Descent

With the fast development of big data, learning the optimal decision rule by recursively updating it and making online decisions has been easier than before. We study the online statistical inference of model parameters in a contextual bandit framework of sequential decision-making. We propose a general framework for an online and adaptive data collection environment that can update decision rules via weighted stochastic gradient descent. We allow different weighting schemes of the stochastic gradient and establish the asymptotic normality of the parameter estimator. Our proposed estimator significantly improves the asymptotic efficiency over the previous averaged SGD approach via inverse probability weights. We also conduct an optimality analysis on the weights in a linear regression setting. We provide a Bahadur representation of the proposed estimator and show that the remainder term in the Bahadur representation entails a slower convergence rate compared to classical SGD due to the adaptive data collection.

stat.ML↗

arXiv2025

Orthus: Autoregressive Interleaved Image-Text Generation with Modality-Specific Heads

We introduce Orthus, an autoregressive (AR) transformer that excels in generating images given textual prompts, answering questions based on visual inputs, and even crafting lengthy image-text interleaved contents. Unlike prior arts on unified multimodal modeling, Orthus simultaneously copes with discrete text tokens and continuous image features under the AR modeling principle. The continuous treatment of visual signals minimizes the information loss for both image understanding and generation while the fully AR formulation renders the characterization of the correlation between modalities straightforward. The key mechanism enabling Orthus to leverage these advantages lies in its modality-specific heads -- one regular language modeling (LM) head predicts discrete text tokens and one diffusion head generates continuous image features conditioning on the output of the backbone. We devise an efficient strategy for building Orthus -- by substituting the Vector Quantization (VQ) operation in the existing unified AR model with a soft alternative, introducing a diffusion head, and tuning the added modules to reconstruct images, we can create an Orthus-base model effortlessly (e.g., within mere 72 A100 GPU hours). Orthus-base can further embrace post-training to better model interleaved images and texts. Empirically, Orthus surpasses competing baselines including Show-o and Chameleon across standard benchmarks, achieving a GenEval score of 0.58 and an MME-P score of 1265.8 using 7B parameters. Orthus also shows exceptional mixed-modality generation capabilities, reflecting the potential for handling intricate practical generation tasks.

cs.CV↗

arXiv2025

RealVVT: Towards Photorealistic Video Virtual Try-on via Spatio-Temporal Consistency

Virtual try-on has emerged as a pivotal task at the intersection of computer vision and fashion, aimed at digitally simulating how clothing items fit on the human body. Despite notable progress in single-image virtual try-on (VTO), current methodologies often struggle to preserve a consistent and authentic appearance of clothing across extended video sequences. This challenge arises from the complexities of capturing dynamic human pose and maintaining target clothing characteristics. We leverage pre-existing video foundation models to introduce RealVVT, a photoRealistic Video Virtual Try-on framework tailored to bolster stability and realism within dynamic video contexts. Our methodology encompasses a Clothing & Temporal Consistency strategy, an Agnostic-guided Attention Focus Loss mechanism to ensure spatial consistency, and a Pose-guided Long Video VTO technique adept at handling extended video sequences.Extensive experiments across various datasets confirms that our approach outperforms existing state-of-the-art models in both single-image and video VTO tasks, offering a viable solution for practical applications within the realms of fashion e-commerce and virtual fitting environments.

cs.CV↗

arXiv2024

Enhancing Decision Transformer with Diffusion-Based Trajectory Branch Generation

Decision Transformer (DT) can learn effective policy from offline datasets by converting the offline reinforcement learning (RL) into a supervised sequence modeling task, where the trajectory elements are generated auto-regressively conditioned on the return-to-go (RTG).However, the sequence modeling learning approach tends to learn policies that converge on the sub-optimal trajectories within the dataset, for lack of bridging data to move to better trajectories, even if the condition is set to the highest RTG.To address this issue, we introduce Diffusion-Based Trajectory Branch Generation (BG), which expands the trajectories of the dataset with branches generated by a diffusion model.The trajectory branch is generated based on the segment of the trajectory within the dataset, and leads to trajectories with higher returns.We concatenate the generated branch with the trajectory segment as an expansion of the trajectory.After expanding, DT has more opportunities to learn policies to move to better trajectories, preventing it from converging to the sub-optimal trajectories.Empirically, after processing with BG, DT outperforms state-of-the-art sequence modeling methods on D4RL benchmark, demonstrating the effectiveness of adding branches to the dataset without further modifications.

cs.LG↗

arXiv2024

Acceleration for Deep Reinforcement Learning using Parallel and Distributed Computing: A Survey

Deep reinforcement learning has led to dramatic breakthroughs in the field of artificial intelligence for the past few years. As the amount of rollout experience data and the size of neural networks for deep reinforcement learning have grown continuously, handling the training process and reducing the time consumption using parallel and distributed computing is becoming an urgent and essential desire. In this paper, we perform a broad and thorough investigation on training acceleration methodologies for deep reinforcement learning based on parallel and distributed computing, providing a comprehensive survey in this field with state-of-the-art methods and pointers to core references. In particular, a taxonomy of literature is provided, along with a discussion of emerging topics and open issues. This incorporates learning system architectures, simulation parallelism, computing parallelism, distributed synchronization mechanisms, and deep evolutionary reinforcement learning. Further, we compare 16 current open-source libraries and platforms with criteria of facilitating rapid development. Finally, we extrapolate future directions that deserve further research.

cs.LG↗

arXiv2024

Uncertainty Quantification of Data Shapley via Statistical Inference

As data plays an increasingly pivotal role in decision-making, the emergence of data markets underscores the growing importance of data valuation. Within the machine learning landscape, Data Shapley stands out as a widely embraced method for data valuation. However, a limitation of Data Shapley is its assumption of a fixed dataset, contrasting with the dynamic nature of real-world applications where data constantly evolves and expands. This paper establishes the relationship between Data Shapley and infinite-order U-statistics and addresses this limitation by quantifying the uncertainty of Data Shapley with changes in data distribution from the perspective of U-statistics. We make statistical inferences on data valuation to obtain confidence intervals for the estimations. We construct two different algorithms to estimate this uncertainty and provide recommendations for their applicable situations. We also conduct a series of experiments on various datasets to verify asymptotic normality and propose a practical trading scenario enabled by this method.

stat.ML↗

arXiv2024

Broadband transmissive polarization rotator by gradiently-twisted $α$-MoO$_3$

Polarization engineering has been proven to enhance the capabilities of light manipulation and thus facilitate the development of integrated photonic devices. In this study, we introduce a polarization rotator based on gradiently-twisted $α$-MoO$_3$ thin film, that works for the mid infrared range and functions in a transmission mode. To be specific, the proposed device is constructed by gradiently-twisted $α$-MoO$_3$ multilayers with a subwavelength thickness of only 5 microns, namely, one-third of the working wavelength. Our analytical calculation demonstrates the efficacy of this subwavelength thin film rotator in converting a linearly polarized wave into its orthogonal counterpart, thanks to its chiral nature. The twisted $α$-MoO$_3$ multilayers exhibit the capability to significantly manipulate dispersion characteristics while maintaining low optical losses, thereby enabling a wide bandwidth exceeding 2.5 THz with a polarization ratio surpassing 17 dB. Moreover, the operational frequency can be adjusted across a 3.4 THz range by altering the incident angle of the incident waves. This adaptable design, characterized by its polarization versatility, can be customized to suit practical applications within wireless communication, radar systems, optical switching, and imaging technologies.

physics.optics↗

arXiv2024

Tunable double notch filter on thin-film lithium niobate platform

Tunable optical filter at the chip scale plays a crucial role in fulfilling the need for the reconfigurability in channel routing, optical switching, and wavelength division multiplexing systems. In this letter, we propose a tunable double notch filter on thin-film lithium niobate using dual micro-ring architecture. This unique integrated filter is essential for complex photonic integrated circuits, along with multiple channels and various frequency spacing. With only one loaded voltage, the device demonstrates a wide frequency spacing tunability from 16.1 GHz to 89.9 GHz by reversely tunning the resonances of the two micro-rings while the center wavelength between the two resonances remains unaltered. Moreover, by utilizing the pronounced electro-optic properties of lithium niobate, associated with the tight light confinement nanophotonic waveguides, the device demonstrates a spacing tunability of 0.82 GHz/V and a contrast of 10~16 dB. In addition, the device has an ultracompact footprint of 0.0248 mm2.

physics.optics↗

arXiv2023

Harvesting correlations from vacuum quantum fields in the presence of a reflecting boundary

We explore correlations harvesting by two static detectors locally interacting with vacuum massless scalar fields in the presence of an infinite perfectly reflecting boundary. We study the phenomena of mutual information harvesting and entanglement harvesting for two detector-boundary alignments, i.e., parallel-to-boundary and vertical-to-boundary alignments. Our results show that the presence of the boundary generally inhibits mutual information harvesting relative to that in flat spacetime without any boundaries. In contrast, the boundary may play a doubled-edged role in entanglement harvesting, i.e., inhibiting entanglement harvesting in the near zone of the boundary while assisting it in the far zone of the boundary. Moreover, there exists an optimal detector energy gap difference between two nonidentical detectors that makes such detectors advantageous in correlations harvesting as long as the interdetector separation is large enough. The value of the optimal detector energy gap difference depends on both the interdetector separation and the detector-to-boundary distance. A comparison of the correlations harvesting in two different alignments shows that although correlations harvesting share qualitatively the same properties, they also display quantitative differences in that the detectors in vertical-to-boundary alignment always harvest comparatively more mutual information than the parallel-to-boundary ones, while they harvest comparatively more entanglement only near the boundary.

quant-ph↗

arXiv2023

2D-Shapley: A Framework for Fragmented Data Valuation

Data valuation -- quantifying the contribution of individual data sources to certain predictive behaviors of a model -- is of great importance to enhancing the transparency of machine learning and designing incentive systems for data sharing. Existing work has focused on evaluating data sources with the shared feature or sample space. How to valuate fragmented data sources of which each only contains partial features and samples remains an open question. We start by presenting a method to calculate the counterfactual of removing a fragment from the aggregated data matrix. Based on the counterfactual calculation, we further propose 2D-Shapley, a theoretical framework for fragmented data valuation that uniquely satisfies some appealing axioms in the fragmented data context. 2D-Shapley empowers a range of new use cases, such as selecting useful data fragments, providing interpretation for sample-wise data values, and fine-grained data issue diagnosis.

cs.LG↗