arXiv subjects

Qi Zhang

Publications and source records attributed to Qi Zhang.

At least 469 records · Page 26Linked to original sources

arXiv2025

Tumor-associated CD19$^+$ macrophages induce immunosuppressive microenvironment in hepatocellular carcinoma

Tumor-associated macrophages are a key component that contributes to the immunosuppressive microenvironment in human cancers. However, therapeutic targeting of macrophages has been a challenge in clinic due to the limited understanding of their heterogeneous subpopulations and distinct functions. Here, we identify a unique and clinically relevant CD19$^+$ subpopulation of macrophages that is enriched in many types of cancer, particularly in hepatocellular carcinoma (HCC). The CD19$^+$ macrophages exhibit increased levels of PD-L1 and CD73, enhanced mitochondrial oxidation, and compromised phagocytosis, indicating their immunosuppressive functions. Targeting CD19$^+$ macrophages with anti-CD19 chimeric antigen receptor T (CAR-T) cells inhibited HCC tumor growth. We identify PAX5 as a primary driver of up-regulated mitochondrial biogenesis in CD19$^+$ macrophages, which depletes cytoplasmic Ca$^{2+}$, leading to lysosomal deficiency and consequent accumulation of CD73 and PD-L1. Inhibiting CD73 or mitochondrial oxidation enhanced the efficacy of immune checkpoint blockade therapy in treating HCC, suggesting great promise for CD19$^+$ macrophage-targeting therapeutics.

q-bio.CB↗

arXiv2025

Mitigating Ambiguities in 3D Classification with Gaussian Splatting

3D classification with point cloud input is a fundamental problem in 3D vision. However, due to the discrete nature and the insufficient material description of point cloud representations, there are ambiguities in distinguishing wire-like and flat surfaces, as well as transparent or reflective objects. To address these issues, we propose Gaussian Splatting (GS) point cloud-based 3D classification. We find that the scale and rotation coefficients in the GS point cloud help characterize surface types. Specifically, wire-like surfaces consist of multiple slender Gaussian ellipsoids, while flat surfaces are composed of a few flat Gaussian ellipsoids. Additionally, the opacity in the GS point cloud represents the transparency characteristics of objects. As a result, ambiguities in point cloud-based 3D classification can be mitigated utilizing GS point cloud as input. To verify the effectiveness of GS point cloud input, we construct the first real-world GS point cloud dataset in the community, which includes 20 categories with 200 objects in each category. Experiments not only validate the superiority of GS point cloud input, especially in distinguishing ambiguous objects, but also demonstrate the generalization ability across different classification methods.

cs.CV↗

arXiv2025

Mitigating Object Hallucinations in MLLMs via Multi-Frequency Perturbations

Recently, multimodal large language models (MLLMs) have demonstrated remarkable performance in visual-language tasks. However, the authenticity of the responses generated by MLLMs is often compromised by object hallucinations. We identify that a key cause of these hallucinations is the model's over-susceptibility to specific image frequency features in detecting objects. In this paper, we introduce Multi-Frequency Perturbations (MFP), a simple, cost-effective, and pluggable method that leverages both low-frequency and high-frequency features of images to perturb visual feature representations and explicitly suppress redundant frequency-domain features during inference, thereby mitigating hallucinations. Experimental results demonstrate that our method significantly mitigates object hallucinations across various model architectures. Furthermore, as a training-time method, MFP can be combined with inference-time methods to achieve state-of-the-art performance on the CHAIR benchmark.

cs.CV↗

arXiv2025

Deep Reinforcement Learning-based Video-Haptic Radio Resource Slicing in Tactile Internet

Enabling video-haptic radio resource slicing in the Tactile Internet requires a sophisticated strategy to meet the distinct requirements of video and haptic data, ensure their synchronized transmission, and address the stringent latency demands of haptic feedback. This paper introduces a Deep Reinforcement Learning-based radio resource slicing framework that addresses video-haptic teleoperation challenges by dynamically balancing radio resources between the video and haptic modalities. The proposed framework employs a refined reward function that considers latency, packet loss, data rate, and the synchronization requirements of both modalities to optimize resource allocation. By catering to the specific service requirements of video-haptic teleoperation, the proposed framework achieves up to a 25% increase in user satisfaction over existing methods, while maintaining effective resource slicing with execution intervals up to 50 ms.

eess.SP↗

arXiv2025

U2AD: Uncertainty-based Unsupervised Anomaly Detection Framework for Detecting T2 Hyperintensity in MRI Spinal Cord

T2 hyperintensities in spinal cord MR images are crucial biomarkers for conditions such as degenerative cervical myelopathy. However, current clinical diagnoses primarily rely on manual evaluation. Deep learning methods have shown promise in lesion detection, but most supervised approaches are heavily dependent on large, annotated datasets. Unsupervised anomaly detection (UAD) offers a compelling alternative by eliminating the need for abnormal data annotations. However, existing UAD methods rely on curated normal datasets and their performance frequently deteriorates when applied to clinical datasets due to domain shifts. We propose an Uncertainty-based Unsupervised Anomaly Detection framework, termed U2AD, to address these limitations. Unlike traditional methods, U2AD is designed to be trained and tested within the same clinical dataset, following a "mask-and-reconstruction" paradigm built on a Vision Transformer-based architecture. We introduce an uncertainty-guided masking strategy to resolve task conflicts between normal reconstruction and anomaly detection to achieve an optimal balance. Specifically, we employ a Monte-Carlo sampling technique to estimate reconstruction uncertainty mappings during training. By iteratively optimizing reconstruction training under the guidance of both epistemic and aleatoric uncertainty, U2AD reduces overall reconstruction variance while emphasizing regions. Experimental results demonstrate that U2AD outperforms existing supervised and unsupervised methods in patient-level identification and segment-level localization tasks. This framework establishes a new benchmark for incorporating uncertainty guidance into UAD, highlighting its clinical utility in addressing domain shifts and task conflicts in medical image anomaly detection. Our code is available: https://github.com/zhibaishouheilab/U2AD

eess.IV↗

arXiv2025

MVD-HuGaS: Human Gaussians from a Single Image via 3D Human Multi-view Diffusion Prior

3D human reconstruction from a single image is a challenging problem and has been exclusively studied in the literature. Recently, some methods have resorted to diffusion models for guidance, optimizing a 3D representation via Score Distillation Sampling(SDS) or generating one back-view image for facilitating reconstruction. However, these methods tend to produce unsatisfactory artifacts (\textit{e.g.} flattened human structure or over-smoothing results caused by inconsistent priors from multiple views) and struggle with real-world generalization in the wild. In this work, we present \emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via a multi-view human diffusion model. We first generate multi-view images from the single reference image with an enhanced multi-view diffusion model, which is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry priors and human structure priors. To infer accurate camera poses from the sparse generated multi-view images for reconstruction, an alignment module is introduced to facilitate joint optimization of 3D Gaussians and camera poses. Furthermore, we propose a depth-based Facial Distortion Mitigation module to refine the generated facial regions, thereby improving the overall fidelity of the reconstruction.Finally, leveraging the refined multi-view images, along with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the target human for high-fidelity free-view renderings. Extensive experiments on Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves state-of-the-art performance on single-view 3D human rendering.

cs.CV↗

arXiv2025

Seedream 2.0: A Native Chinese-English Bilingual Image Generation Foundation Model

Rapid advancement of diffusion models has catalyzed remarkable progress in the field of image generation. However, prevalent models such as Flux, SD3.5 and Midjourney, still grapple with issues like model bias, limited text rendering capabilities, and insufficient understanding of Chinese cultural nuances. To address these limitations, we present Seedream 2.0, a native Chinese-English bilingual image generation foundation model that excels across diverse dimensions, which adeptly manages text prompt in both Chinese and English, supporting bilingual image generation and text rendering. We develop a powerful data system that facilitates knowledge integration, and a caption system that balances the accuracy and richness for image description. Particularly, Seedream is integrated with a self-developed bilingual large language model as a text encoder, allowing it to learn native knowledge directly from massive data. This enable it to generate high-fidelity images with accurate cultural nuances and aesthetic expressions described in either Chinese or English. Beside, Glyph-Aligned ByT5 is applied for flexible character-level text rendering, while a Scaled ROPE generalizes well to untrained resolutions. Multi-phase post-training optimizations, including SFT and RLHF iterations, further improve the overall capability. Through extensive experimentation, we demonstrate that Seedream 2.0 achieves state-of-the-art performance across multiple aspects, including prompt-following, aesthetics, text rendering, and structural correctness. Furthermore, Seedream 2.0 has been optimized through multiple RLHF iterations to closely align its output with human preferences, as revealed by its outstanding ELO score. In addition, it can be readily adapted to an instruction-based image editing model, such as SeedEdit, with strong editing capability that balances instruction-following and image consistency.

cs.CV↗

arXiv2025

Convergence Guarantees for RMSProp and Adam in Generalized-smooth Non-convex Optimization with Affine Noise Variance

This paper provides the first tight convergence analyses for RMSProp and Adam in non-convex optimization under the most relaxed assumptions of coordinate-wise generalized smoothness and affine noise variance. We first analyze RMSProp, which is a special case of Adam with adaptive learning rates but without first-order momentum. Specifically, to solve the challenges due to dependence among adaptive update, unbounded gradient estimate and Lipschitz constant, we demonstrate that the first-order term in the descent lemma converges and its denominator is upper bounded by a function of gradient norm. Based on this result, we show that RMSProp with proper hyperparameters converges to an $ε$-stationary point with an iteration complexity of $\mathcal O(ε^{-4})$. We then generalize our analysis to Adam, where the additional challenge is due to a mismatch between the gradient and first-order momentum. We develop a new upper bound on the first-order term in the descent lemma, which is also a function of the gradient norm. We show that Adam with proper hyperparameters converges to an $ε$-stationary point with an iteration complexity of $\mathcal O(ε^{-4})$. Our complexity results for both RMSProp and Adam match with the complexity lower bound established in \cite{arjevani2023lower}.

stat.ML↗

arXiv2025

MGDA Converges under Generalized Smoothness, Provably

Multi-objective optimization (MOO) is receiving more attention in various fields such as multi-task learning. Recent works provide some effective algorithms with theoretical analysis but they are limited by the standard $L$-smooth or bounded-gradient assumptions, which typically do not hold for neural networks, such as Long short-term memory (LSTM) models and Transformers. In this paper, we study a more general and realistic class of generalized $\ell$-smooth loss functions, where $\ell$ is a general non-decreasing function of gradient norm. We revisit and analyze the fundamental multiple gradient descent algorithm (MGDA) and its stochastic version with double sampling for solving the generalized $\ell$-smooth MOO problems, which approximate the conflict-avoidant (CA) direction that maximizes the minimum improvement among objectives. We provide a comprehensive convergence analysis of these algorithms and show that they converge to an $ε$-accurate Pareto stationary point with a guaranteed $ε$-level average CA distance (i.e., the gap between the updating direction and the CA direction) over all iterations, where totally $\mathcal{O}(ε^{-2})$ and $\mathcal{O}(ε^{-4})$ samples are needed for deterministic and stochastic settings, respectively. We prove that they can also guarantee a tighter $ε$-level CA distance in each iteration using more samples. Moreover, we analyze an efficient variant of MGDA named MGDA-FA using only $\mathcal{O}(1)$ time and space, while achieving the same performance guarantee as MGDA.

cs.LG↗

arXiv2025

Column generation for multistage stochastic mixed-integer nonlinear programs with discrete state variables

Stochastic programming provides a natural framework for modeling sequential optimization problems under uncertainty; however, the efficient solution of large-scale multistage stochastic programs remains a challenge, especially in the presence of discrete decisions and nonlinearities. In this work, we consider multistage stochastic mixed-integer nonlinear programs (MINLPs) with discrete state variables, which exhibit a decomposable structure that allows its solution using a column generation approach. Following a Dantzig-Wolfe reformulation, we apply column generation such that each pricing subproblem is an MINLP of much smaller size, making it more amenable to global MINLP solvers. We further propose a method for generating additional columns that satisfy the nonanticipativity constraints, leading to significantly improved convergence and optimal or near-optimal solutions for many large-scale instances in a reasonable computation time. The effectiveness of the tailored column generation algorithm is demonstrated via computational case studies on a multistage blending problem and a problem involving the routing of mobile generators in a power distribution network.

math.OC↗

arXiv2025

Better Process Supervision with Bi-directional Rewarding Signals

Process supervision, i.e., evaluating each step, is critical for complex large language model (LLM) reasoning and test-time searching with increased inference compute. Existing approaches, represented by process reward models (PRMs), primarily focus on rewarding signals up to the current step, exhibiting a one-directional nature and lacking a mechanism to model the distance to the final target. To address this problem, we draw inspiration from the A* algorithm, which states that an effective supervisory signal should simultaneously consider the incurred cost and the estimated cost for reaching the target. Building on this key insight, we introduce BiRM, a novel process supervision model that not only evaluates the correctness of previous steps but also models the probability of future success. We conduct extensive experiments on mathematical reasoning tasks and demonstrate that BiRM provides more precise evaluations of LLM reasoning steps, achieving an improvement of 3.1% on Gaokao2023 over PRM under the Best-of-N sampling method. Besides, in search-based strategies, BiRM provides more comprehensive guidance and outperforms ORM by 5.0% and PRM by 3.8% respectively on MATH-500.

cs.CL↗

arXiv2025

Projection Head is Secretly an Information Bottleneck

Recently, contrastive learning has risen to be a promising paradigm for extracting meaningful data representations. Among various special designs, adding a projection head on top of the encoder during training and removing it for downstream tasks has proven to significantly enhance the performance of contrastive learning. However, despite its empirical success, the underlying mechanism of the projection head remains under-explored. In this paper, we develop an in-depth theoretical understanding of the projection head from the information-theoretic perspective. By establishing the theoretical guarantees on the downstream performance of the features before the projector, we reveal that an effective projector should act as an information bottleneck, filtering out the information irrelevant to the contrastive objective. Based on theoretical insights, we introduce modifications to projectors with training and structural regularizations. Empirically, our methods exhibit consistent improvement in the downstream performance across various real-world datasets, including CIFAR-10, CIFAR-100, and ImageNet-100. We believe our theoretical understanding on the role of the projection head will inspire more principled and advanced designs in this field. Code is available at https://github.com/PKU-ML/Projector_Theory.

cs.LG↗

arXiv2025

MindSimulator: Exploring Brain Concept Localization via Synthetic FMRI

Concept-selective regions within the human cerebral cortex exhibit significant activation in response to specific visual stimuli associated with particular concepts. Precisely localizing these regions stands as a crucial long-term goal in neuroscience to grasp essential brain functions and mechanisms. Conventional experiment-driven approaches hinge on manually constructed visual stimulus collections and corresponding brain activity recordings, constraining the support and coverage of concept localization. Additionally, these stimuli often consist of concept objects in unnatural contexts and are potentially biased by subjective preferences, thus prompting concerns about the validity and generalizability of the identified regions. To address these limitations, we propose a data-driven exploration approach. By synthesizing extensive brain activity recordings, we statistically localize various concept-selective regions. Our proposed MindSimulator leverages advanced generative technologies to learn the probability distribution of brain activity conditioned on concept-oriented visual stimuli. This enables the creation of simulated brain recordings that reflect real neural response patterns. Using the synthetic recordings, we successfully localize several well-studied concept-selective regions and validate them against empirical findings, achieving promising prediction accuracy. The feasibility opens avenues for exploring novel concept-selective regions and provides prior hypotheses for future neuroscience research.

q-bio.NC↗

arXiv2025

PASemiQA: Plan-Assisted Agent for Question Answering on Semi-Structured Data with Text and Relational Information

Large language models (LLMs) have shown impressive abilities in answering questions across various domains, but they often encounter hallucination issues on questions that require professional and up-to-date knowledge. To address this limitation, retrieval-augmented generation (RAG) techniques have been proposed, which retrieve relevant information from external sources to inform their responses. However, existing RAG methods typically focus on a single type of external data, such as vectorized text database or knowledge graphs, and cannot well handle real-world questions on semi-structured data containing both text and relational information. To bridge this gap, we introduce PASemiQA, a novel approach that jointly leverages text and relational information in semi-structured data to answer questions. PASemiQA first generates a plan to identify relevant text and relational information to answer the question in semi-structured data, and then uses an LLM agent to traverse the semi-structured data and extract necessary information. Our empirical results demonstrate the effectiveness of PASemiQA across different semi-structured datasets from various domains, showcasing its potential to improve the accuracy and reliability of question answering systems on semi-structured data.

cs.CL↗

arXiv2025

VEM: Environment-Free Exploration for Training GUI Agent with Value Environment Model

Training Vision-Language Models (VLMs) for Graphical User Interfaces (GUI) agents via Reinforcement Learning (RL) faces critical challenges: environment-based RL requires costly interactions, while environment-free methods struggle with distribution shift and reward generalization. We propose an environment-free RL framework that decouples value estimation from policy optimization by leveraging a pretrained Value Environment Model (VEM). VEM predicts state-action values directly from offline data, distilling human-like priors about GUI interaction outcomes without requiring next-state prediction or environmental feedback. This avoids compounding errors and enhances resilience to UI changes by focusing on semantic reasoning (e.g., Does this action advance the user's goal?). The framework operates in two stages: (1) pretraining VEM to estimate long-term action utilities and (2) guiding policy exploration with frozen VEM signals, enabling layout-agnostic GUI automation. Evaluated on Android-in-the-Wild benchmarks, VEM achieves state-of-the-art performance in both offline and online settings, outperforming environment-free baselines significantly and matching environment-based approaches without interaction costs. Importantly, VEM demonstrates that semantic-aware value estimation can achieve comparable performance with online-trained methods.

cs.LG↗

arXiv2025

Distill Not Only Data but Also Rewards: Can Smaller Language Models Surpass Larger Ones?

Distilling large language models (LLMs) typically involves transferring the teacher model's responses through supervised fine-tuning (SFT). However, this approach neglects the potential to distill both data (output content) and reward signals (quality evaluations). Extracting reliable reward signals directly from teacher models is challenging, as LLMs are optimized for generation rather than evaluation, often resulting in biased or inconsistent assessments. To address this limitation, we propose a novel distillation pipeline that transfers both responses and rewards. Our method generates pseudo-rewards through a self-supervised mechanism that leverages the inherent structure of both teacher and student responses, enabling reward learning without explicit external evaluation. The reward model subsequently guides reinforcement learning (RL), allowing iterative refinement of the student model after an SFT warm-up phase. Experiments on GSM8K and MMLU-PRO demonstrate that our method consistently outperforms traditional SFT-based approaches, enabling student models to surpass the performance of their teachers. This work highlights the potential for scalable, efficient distillation through structured self-supervised reward learning, reducing dependence on external reward supervision.

cs.CL↗

arXiv2025

SHRINK: Data Compression by Semantic Extraction and Residuals Encoding

The distributed data infrastructure in Internet of Things (IoT) ecosystems requires efficient data-series compression methods, along with the ability to feed different accuracy demands. However, the compression performance of existing compression methods degrades sharply when calling for ultra-accurate data recovery. In this paper, we introduce SHRINK, a novel highly accurate data compression method that offers a higher compression ratio and also lower runtime than prior compressors. SHRINK extracts data semantics in the form of linear segments to construct a compact knowledge base, using a dynamic error threshold that it adapts to data characteristics. Then, it captures the remaining data details as residuals to support lossy compression at diverse resolutions as well as lossless compression. As SHRINK identifies repeated semantics, its compression ratio increases with data size. Our experimental evaluation demonstrates that SHRINK outperforms state-of-art methods with an up to threefold improvement in compression ratio.

cs.DC↗

arXiv2025

Generating with Fairness: A Modality-Diffused Counterfactual Framework for Incomplete Multimodal Recommendations

Incomplete scenario is a prevalent, practical, yet challenging setting in Multimodal Recommendations (MMRec), where some item modalities are missing due to various factors. Recently, a few efforts have sought to improve the recommendation accuracy by exploring generic structures from incomplete data. However, two significant gaps persist: 1) the difficulty in accurately generating missing data due to the limited ability to capture modality distributions; and 2) the critical but overlooked visibility bias, where items with missing modalities are more likely to be disregarded due to the prioritization of items' multimodal data over user preference alignment. This bias raises serious concerns about the fair treatment of items. To bridge these two gaps, we propose a novel Modality-Diffused Counterfactual (MoDiCF) framework for incomplete multimodal recommendations. MoDiCF features two key modules: a novel modality-diffused data completion module and a new counterfactual multimodal recommendation module. The former, equipped with a particularly designed multimodal generative framework, accurately generates and iteratively refines missing data from learned modality-specific distribution spaces. The latter, grounded in the causal perspective, effectively mitigates the negative causal effects of visibility bias and thus assures fairness in recommendations. Both modules work collaboratively to address the two aforementioned significant gaps for generating more accurate and fair results. Extensive experiments on three real-world datasets demonstrate the superior performance of MoDiCF in terms of both recommendation accuracy and fairness. The code and processed datasets are released at https://github.com/JinLi-i/MoDiCF.

cs.IR↗