arXiv subjects

Jing Jia

Publications and source records attributed to Jing Jia.

9 recordsLinked to original sources

arXiv2026

Couple to Control: Joint Initial Noise Design in Diffusion Models

Diffusion models typically generate image batches from independent Gaussian initial noises. We argue that this independence assumption is only one choice within a broader class of valid joint noise designs. Instead, one can specify a coupling of the initial noises: each noise remains marginally standard Gaussian, so the pretrained diffusion model receives the same single-sample input distribution, while the dependence across samples is chosen by design. This reframes initial-noise control from selecting or optimizing individual seeds to designing the dependence structure of a multi-sample gallery. This view gives a general framework for initial-noise design, covering several existing methods as special cases and leading naturally to new coupled-noise constructions. Coupled noise can improve generation on its own without adding sampling cost, and it is flexible enough to serve as a structured initialization for optimization-based pipelines when additional computation is available. Empirically, repulsive Gaussian coupling improves gallery diversity on SD1.5, SDXL, and SD3 while largely preserving prompt alignment and image quality. It matches or outperforms recent test-time noise-optimization baselines on several diversity metrics at the same sampling cost as independent generation. Subspace couplings also support fixed-object background generation, producing diverse, natural backgrounds compared with specialized inpainting baselines, with a tunable trade-off in foreground fidelity.

cs.LG↗

arXiv2026

Weak Diffusion Priors Can Still Achieve Strong Inverse-Problem Performance

Can a diffusion model trained on bedrooms recover human faces? Diffusion models are widely used as priors for inverse problems, but standard approaches usually assume a high-fidelity model trained on data that closely match the unknown signal. In practice, one often must use a mismatched or low-fidelity diffusion prior. Surprisingly, these weak priors often perform nearly as well as full-strength, in-domain baselines. We study when and why inverse solvers are robust to weak diffusion priors. Through extensive experiments, we find that weak priors succeed when measurements are highly informative (e.g., many observed pixels), and we identify regimes where they fail. To explain this behavior, we combine Bayesian-consistency theory with local-correlation analysis: the theory gives conditions under which high-dimensional measurements make the posterior concentrate near the true signal, while the correlation analysis shows that weak and stronger natural-image priors can share similar local spatial structure. These results provide a principled justification on when weak diffusion priors can be used reliably. Code is available at https://github.com/jjia131/weak-diffusion-priors-inverse-problem.

cs.LG↗

arXiv2025

Antithetic Noise in Diffusion Models

We systematically study antithetic initial noise in diffusion models, discovering that pairing each noise sample with its negation consistently produces strong negative correlation. This universal phenomenon holds across datasets, model architectures, conditional and unconditional sampling, and even other generative models such as VAEs and Normalizing Flows. To explain it, we combine experiments and theory and propose a \textit{symmetry conjecture} that the learned score function is approximately affine antisymmetric (odd symmetry up to a constant shift), supported by empirical evidence. This negative correlation leads to substantially more reliable uncertainty quantification with up to $90\%$ narrower confidence intervals. We demonstrate these gains on tasks including estimating pixel-wise statistics and evaluating diffusion inverse solvers. We also provide extensions with randomized quasi-Monte Carlo noise designs for uncertainty quantification, and explore additional applications of the antithetic noise design to improve image editing and generation diversity. Our framework is training-free, model-agnostic, and adds no runtime overhead. Code is available at https://github.com/jjia131/Antithetic-Noise-in-Diffusion-Models-page.

cs.LG↗

arXiv2025

Privacy-Preserving Socialized Recommendation based on Multi-View Clustering in a Cloud Environment

Recommendation as a service has improved the quality of our lives and plays a significant role in variant aspects. However, the preference of users may reveal some sensitive information, so that the protection of privacy is required. In this paper, we propose a privacy-preserving, socialized, recommendation protocol that introduces information collected from online social networks to enhance the quality of the recommendation. The proposed scheme can calculate the similarity between users to determine their potential relationships and interests, and it also can protect the users' privacy from leaking to an untrusted third party. The security analysis and experimental results showed that our proposed scheme provides excellent performance and is feasible for real-world applications.

cs.CR↗

arXiv2025

Hallucinatory Image Tokens: A Training-free EAZY Approach on Detecting and Mitigating Object Hallucinations in LVLMs

Despite their remarkable potential, Large Vision-Language Models (LVLMs) still face challenges with object hallucination, a problem where their generated outputs mistakenly incorporate objects that do not actually exist. Although most works focus on addressing this issue within the language-model backbone, our work shifts the focus to the image input source, investigating how specific image tokens contribute to hallucinations. Our analysis reveals a striking finding: a small subset of image tokens with high attention scores are the primary drivers of object hallucination. By removing these hallucinatory image tokens (only 1.5% of all image tokens), the issue can be effectively mitigated. This finding holds consistently across different models and datasets. Building on this insight, we introduce EAZY, a novel, training-free method that automatically identifies and Eliminates hAllucinations by Zeroing out hallucinatorY image tokens. We utilize EAZY for unsupervised object hallucination detection, achieving 15% improvement compared to previous methods. Additionally, EAZY demonstrates remarkable effectiveness in mitigating hallucinations while preserving model utility and seamlessly adapting to various LVLM architectures.

cs.CV↗

arXiv2025

CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation

Diffusion models have emerged as powerful tools for generative tasks, producing high-quality outputs across diverse domains. However, how the generated data responds to the initial noise perturbation in diffusion models remains under-explored, which hinders understanding the controllability of the sampling process. In this work, we first observe an interesting phenomenon: the relationship between the change of generation outputs and the scale of initial noise perturbation is highly linear through the diffusion ODE sampling. Then we provide both theoretical and empirical study to justify this linearity property of this input-output (noise-generation data) relationship. Inspired by these new insights, we propose a novel Controllable and Constrained Sampling method (CCS) together with a new controller algorithm for diffusion models to sample with desired statistical properties while preserving good sample quality. We perform extensive experiments to compare our proposed sampling approach with other methods on both sampling controllability and sampled data quality. Results show that our CCS method achieves more precisely controlled sampling while maintaining superior sample quality and diversity.

cs.LG↗

arXiv2024

Left-handedness with three zero-absorption windows tuned by the incoherent pumping field and inter-dot tunnelings in a GaAs/AlGaAs triple quantum dots system

Left-handedness with three zero-absorption windows is achieved in a triple-quantum dot (TQD) system. With the typical parameters of a GaAs/AlGaAs heterostructure, the simultaneous negative relative electric permittivity and magnetic permeability are obtained by the adjustable incoherent pumping field and two inter-dot tunnelings. What's more, three zero-absorption windows in the left-handedness frequency bands are observed in the TQD system. The left-handedness with zero-absorption in solid state heterostructure may solve not only the challenge in the photonic resonant scheme for left-handed materials (LHMs) but also the application limitation of the negative refractive materials with large amount of absorption.

quant-ph↗

arXiv2023

Privacy-Preserving Hierarchical Anonymization Framework over Encrypted Data

Smart cities, which can monitor the real world and provide smart services in a variety of fields, have improved people's living standards as urbanization has accelerated. However, there are security and privacy concerns because smart city applications collect large amounts of privacy-sensitive information from people and their social circles. Anonymization, which generalizes data and reduces data uniqueness is an important step in preserving the privacy of sensitive information. However, anonymization methods frequently require large datasets and rely on untrusted third parties to collect and manage data, particularly in a cloud environment. In this case, private data leakage remains a critical issue, discouraging users from sharing their data and impeding the advancement of smart city services. This problem can be solved if the computational entity can perform the anonymization process without obtaining the original plain text. This study proposed a hierarchical k-anonymization framework using homomorphic encryption and secret sharing composed of two types of domains. Different computing methods are selected flexibly, and two domains are connected hierarchically to obtain higher-level anonymization results in an efficient manner. The experimental results show that connecting two domains can accelerate the anonymization process, indicating that the proposed secure hierarchical architecture is practical and efficient.

cs.CR↗

arXiv2020

Retrieve Synonymous keywords for Frequent Queries in Sponsored Search in a Data Augmentation Way

In sponsored search, retrieving synonymous keywords is of great importance for accurately targeted advertising. The semantic gap between queries and keywords and the extremely high precision requirements (>= 95\%) are two major challenges to this task. To the best of our knowledge, the problem has not been openly discussed. In an industrial sponsored search system, the retrieved keywords for frequent queries are usually done ahead of time and stored in a lookup table. Considering these results as a seed dataset, we propose a data-augmentation-like framework to improve the synonymous retrieval performance for these frequent queries. This framework comprises two steps: translation-based retrieval and discriminant-based filtering. Firstly, we devise a Trie-based translation model to make a data increment. In this phase, a Bag-of-Core-Words trick is conducted, which increased the data increment's volume 4.2 times while keeping the original precision. Then we use a BERT-based discriminant model to filter out nonsynonymous pairs, which exceeds the traditional feature-driven GBDT model with 11\% absolute AUC improvement. This method has been successfully applied to Baidu's sponsored search system, which has yielded a significant improvement in revenue. In addition, a commercial Chinese dataset containing 500K synonymous pairs with a precision of 95\% is released to the public for paraphrase study (http://ai.baidu.com/broad/subordinate?dataset=paraphrasing).

cs.IR↗