Searcharxiv⌕ Search

arXiv subjects

Hao Chen

Publications and source records attributed to Hao Chen.

At least 253 records · Page 14Linked to original sources
arXiv2026

A Breast Vision Pathology Foundation Model for Real-world Clinical Utility

Pathology foundation models have shown strong retrospective performance, but whether such systems can support clinically relevant use remains unclear. This challenge is particularly important in breast cancer, where pathological assessment serves as the gold standard for diagnosis and guides treatment planning, surgical decision-making and risk stratification across pre-, intra- and post-operative stages. Here we present \textbf{BRAVE}, a breast-adaptive pathology foundation model developed and evaluated using a total resource of 101,638 breast whole-slide images from 32 sources across Asia, Europe and North America. We assessed BRAVE across 34 tasks in 82 cohorts spanning pre-operative biopsy, intra-operative frozen section and post-operative resection, using an evidence chain comprising retrospective benchmarking, clinically challenging scenarios, workflow-oriented clinical impact simulations, prospective observational validation with the thresholds locked in the retrospective cohorts and crossover pathologist-AI interaction studies. Across these settings, BRAVE supported practical roles in the clinical workflow, including safe exclusion of low-risk cases from routine review, AI-assisted second-review rescue of initially missed positives and prioritization of cases for further assessment. In prospective validation across three centres, BRAVE excluded 76.9% of negative biopsy cases (NPV 0.953) and 70.1% of negative frozen-section cases (NPV 0.973), and triaged 78.8% of post-operative subtyping cases as high-confidence clear-cut cases (NPV 1.000). In reader studies, AI assistance improved balanced accuracy from 88.5% to 95.1% (OR 3.14, P<0.001), with better efficiency, confidence and inter-rater agreement. BRAVE-derived scores also independently predicted disease-free survival (adjusted HR 4.79, P<0.001) and overall survival (adjusted HR 8.14, P<0.001).

cs.CV↗
arXiv2026

Understanding and Mitigating Bias Inheritance in LLM-based Data Augmentation on Downstream Tasks

Generating synthetic datasets via large language models (LLMs) has emerged as a promising approach to improve LLM performance. However, LLMs inherently reflect biases in their training data, leading to a critical challenge: when models are trained on synthetic data, they may propagate and amplify the inherent biases that can significantly impact fairness and robustness on downstream tasks-a phenomenon we term bias inheritance. This work presents the first systematic investigation in understanding, analyzing, and mitigating bias inheritance. We fine-tune LLMs with a combined dataset of real and LLM-augmented data with varied bias ratio as the proportion of augmented data. Through systematic experiments across 10 classification and generation tasks, we analyze how 6 different types of biases manifest. Our results indicate that bias inheritance harms downstream task performance in bias directly-related classification and generation tasks. Then, our analysis identifies three key misalignment factors: misalignment of values, group data, and data distributions. Based on these insights, we propose three mitigation strategies: token-based, mask-based, and loss-based approaches, which can work differently on various tasks and bias, indicating the substantial challenges to mitigate bias inheritance. We hope this work can provide insights to the research of LLM data augmentation.

cs.LG↗
arXiv2026

Tutti: Making SSD-Backed KV Cache Practical for Long-Context LLM Serving

LLM serving relies on prefix caching to improve inference performance. As growing contexts push key-value (KV) cache footprint far beyond GPU HBM and CPU DRAM capacity, KV cache is increasingly offloaded to NVMe SSDs. Unfortunately, restoring KV cache from SSDs suffers from poor I/O performance and incurs significant GPU stalls. This is primarily because the fragmented GPU memory layout results in a massive number of tiny random I/Os, rendering the low-parallelism CPU a severe bottleneck even with GPU Direct Storage (GDS), which still relies on CPU intervention to initiate each I/O and thus remains CPU-centric. This paper presents Tutti, an efficient SSD-backed KV caching solution that eliminates CPU intervention from the critical data and I/O control paths between HBM and SSDs. At the core of Tutti is a GPU-centric KV cache object store, in which the CPU is only responsible for asynchronously loading I/O kernels once per layer to the GPU. Tutti saturates NVMe SSD bandwidth and reduces GPU stalls to near zero through the following designs: (i) we provide a GPU-native object abstraction that enables bulk KV cache transfers and management; (ii) we re-architect the GPU storage stack by introducing GPU io_uring to support asynchronous GPU direct object I/O; and (iii) we propose slack-aware I/O scheduling to avoid GPU resource contention. We have implemented Tutti and integrated it to vLLM. Extensive evaluation shows that compared to the state-of-the-art GDS-enabled, SSD-backed LMCache, Tutti reduces TTFT by 78.3% under strict SLO constraints and improves the achievable request rate by 2x. The serving cost is reduced by 27%. Tutti achieves nearly the same inference performance as DRAM-backed LMCache, while providing almost infinite capacity.

cs.OS↗
arXiv2026

On the Minimum Distances of Some Families of Goppa Codes and BCH Codes

Goppa codes form an important class of alternant codes with wide applications in algebraic coding theory and code-based cryptography. Determining the true minimum distance of a Goppa code is a difficult problem. In this paper, we provide a necessary and sufficient criterion for a Goppa code to attain its designed distance $δ=t+1$, where $t$ is the degree of the Goppa polynomial. As applications, we determine the minimum distances of several classes of $q$-ary Goppa codes. In particular, we prove the tightness of the improved lower bound for a class of wild Goppa codes, and extend the family with $G(x)=x^t+A$ from the binary case to arbitrary odd prime powers. We then specialize the criterion to the monomial case $G(x)=x^t$, which is equivalent to primitive BCH codes. This leads to several infinite families of primitive BCH codes with $d=δ$, including the binary codes $\mathbf{C}_{(2,2^m-1,9,1)}$ and $\mathbf{C}_{(2,2^m-1,15,1)}$, the family $\mathbf{C}_{(p,p^p-1,2p+2,1)}$ with an odd prime $p$ and the family $\mathbf{C}_{(q,q^m-1,r\frac{q^m-1}{q-1}+1,1)}$ with $r\mid q-1$. In particular, we prove that the primitive BCH code $\mathbf{C}_{(q,q^m-1,q^t+1,1)}$ has minimum distance $q^t+1$ under the condition $t\mid m$, improving the previously known condition $pt\mid m$.

cs.IT↗
arXiv2026

DRIVE: Modeling Skills at the Reasoning and Interaction Levels for Web Agents under Continual Learning

Web agents require both high-level reasoning (for task decomposition) and low-level interactions (for page elements manipulation) to conduct different tasks. However, these knowledge types differ fundamentally: reasoning knowledge (e.g., booking a flight requires first searching for routes) is abstract and transferable across websites, while interaction knowledge (e.g., clicking the Search button at a specific coordinate on Site A) depends heavily on page-specific contexts. Existing methods store experiences uniformly. This creates a dilemma: abstract representations lose executability on concrete pages, while concrete representations fail to generalize across domains. This entanglement limits capability accumulation: on new websites, agents either fail to recognize reusable task logic due to surface-level differences or attempt infeasible actions from outdated page structures. To disentangle them, we propose DRIVE, a dual-level skill modeling framework separating historical experience into natural language reasoning skills, which capture transferable task logic, and programmatic interaction skills, grounding abstract actions to executable operations. A scene-aware coordination mechanism adaptively retrieves and invokes these dual-level skills based on task semantics. DRIVE also uses skill-level reflection to identify hierarchy-specific failure modes, enabling targeted skill library expansion and refinement. Experiments across five WebArena domains show DRIVE attains an average task success rate of 52.8%, exceeding the skill-free baseline by 7.3 percentage points. Further ablations show reasoning and interaction skills provide distinct, complementary benefits, supporting separation of transferable task logic from executable page-level operations.

cs.AI↗
arXiv2026

On the Minimum Distances of Some Families of BCH Codes

BCH codes form an important class of cyclic codes, which have applications in communication and data storage systems. Although the BCH bound provides a lower bound on the minimum distance of BCH codes, determining the true minimum distances of BCH codes is a very challenging problem. In this paper, we settle the minimum distances of a number of infinite families of narrow-sense BCH codes. By explicitly constructing the locator polynomials for minimum weight codewords, we obtain many families of primitive and non-primitive BCH codes with $d=δ$, where $d$ is the minimum distance of a $q$-ary BCH code of length $n$, designed distance $δ$, and offset $b$, denoted by $\mathbf{C}_{(q, n, δ, b)}$. For primitive BCH codes, we obtain infinite families of BCH codes over $\mathbb{F}_3$ and $\mathbb{F}_4$ satisfying $d=δ$, where $δ\in \{5,6,7,8\}$. Moreover, we construct several infinite families of $q$-ary BCH codes with $d=δ$, where $2 \le δ\le q-1$. For $δ=q^t+1$, we prove that the BCH code $\mathbf{C}_{(q, q^m-1, q^t+1, 1)}$ has $d=δ$ for all $m$ satisfying $m \equiv 0 \pmod{pt}$, where $p$ denotes the characteristic of $\mathbb{F}_q$. In the paper by Ding et al., IEEE Trans. Inf. Theory 61(5): 2351-2356, it was conjectured that the minimum distance of $\mathbf{C}_{(q, q^m-1, q^t+1, 1)}$ is always equal to its Bose distance $d_B$. Our result confirms this conjecture for the case $m \equiv 0 \pmod{pt}$. For non-primitive BCH codes, we construct a family of BCH codes $\mathbf{C}_{(q,\frac{q^p-1}λ,p+1,1)}$ with $d=δ=p+1$, where $p$ is an odd prime, $q=p^e$ with $p \nmid e$ and $λ\mid q-1$.

cs.IT↗
arXiv2026

A Hierarchical Ensemble Inference Pipeline for Robust White Blood Cell Classification Under Domain Shifts

Automated white blood cell (WBC) classification is essential for scalable leukaemia screening. However, real-world deployment is challenged by domain shifts caused by staining protocols, scanner characteristics, and inter-laboratory variability, which often degrade model performance. The White Blood Cell Classification Challenge (WBCBench) at ISBI 2026 aims to advance robust WBC recognition, with a focus on accurately identifying blast cells and other clinically critical rare subtypes. We propose a memory-augmented, hierarchical ensemble pipeline for WBC classification under domain shifts, leveraging a feature bank and a DinoBloom backbone fine-tuned with LoRA. Our three-stage inference hierarchy combines k-nearest neighbors (kNN) retrieval at each level, reducing over-reliance on any single decision. Evaluated on the WBCBench dataset, our method ranks within the top ten by macro F1-score in the final testing phase.

cs.CV↗
arXiv2026

Persistent Fermi Pockets and Robust Electron Pairing in Lightly Doped CuO$_2$ Planes of Cuprate Superconductors

High temperature superconductivity in cuprate superconductors is generally considered to be generated from doping the Mott insulators. The fundamental nature of the doped parent compounds as well as the microscopic origin of electron pairing remain critical issues in understanding the emergence of superconductivity. Here, using high-resolution spatially-resolved laser angle-resolved photoemission spectroscopy, we investigate the intrinsic electronic structures of the CuO$_2$ planes in multilayer cuprates Bi$_2$Sr$_2$Ca$_{n-1}$Cu$_n$O$_{2n+4+δ}$ (n=5$\sim$8). The inner CuO$_2$ planes are well shielded from the disorders and provide a rare and ideal platform to probe the intrinsic electronic phase diagram. We observe well-defined Fermi pockets with hole doping levels as low as 0.007, demonstrating an abrupt transition from the parent Mott insulator to a metallic state upon the introduction of an infinitesimal amount of doping. The innermost CuO$_2$ planes (IP$_0$) display gapless Fermi pockets, while the second innermost planes (IP$_1$) exhibit anisotropic superconducting gaps up to $\sim$33$\,$meV, indicative of robust electron pairing coexisting with strong antiferromagnetic order. Our findings provide a revised framework for understanding the doping-driven transitions and pairing mechanisms in cuprate superconductors.

cond-mat.supr-con↗
arXiv2026

Unlocking the Power of Critical Factors for 3D Visual Geometry Estimation

Feed-forward visual geometry estimation has recently made rapid progress. However, an important gap remains: multi-frame models usually produce better cross-frame consistency, yet they often underperform strong per-frame methods on single-frame accuracy. This observation motivates our systematic investigation into the critical factors driving model performance through rigorous ablation studies, which reveals several key insights: 1) Scaling up data diversity and quality unlocks further performance gains even in state-of-the-art visual geometry estimation methods; 2) Commonly adopted confidence-aware loss and gradient-based loss mechanisms may unintentionally hinder performance; 3) Joint supervision through both per-sequence and per-frame alignment improves results, while local region alignment surprisingly degrades performance. Furthermore, we introduce two enhancements to integrate the advantages of optimization-based methods and high-resolution inputs: a consistency loss function that enforces alignment between depth maps, camera parameters, and point maps, and an efficient architectural design that leverages high-resolution information. We integrate these designs into CARVE, a resolution-enhanced model for feed-forward visual geometry estimation. Experiments on point cloud reconstruction, video depth estimation, and camera pose/intrinsic estimation show that CARVE achieves strong and robust performance across diverse benchmarks.

cs.CV↗
arXiv2026

PrismaDV: Automated Task-Aware Data Unit Test Generation

Data is a central resource for modern enterprises, and data validation is essential for ensuring the reliability of downstream applications. However, existing automated data unit testing frameworks are largely task-agnostic: they validate datasets without considering the semantics and requirements of the code that consumes the data. We present PrismaDV, a compound AI system that analyzes downstream task code together with dataset profiles to identify data access patterns, infer implicit data assumptions, and generate task-aware executable data unit tests. To further adapt the data unit tests over time to specific datasets and downstream tasks, we propose "Selective Informative Feedback for Task Adaptation" (SIFTA), a prompt-optimization framework that leverages the scarce outcomes from the execution of data unit tests and downstream tasks. We evaluate PrismaDV on two new benchmarks spanning 60 tasks across five datasets, where it consistently outperforms both task-agnostic and task-aware baselines in generating unit tests that reflect the end-to-end impact of data errors. Furthermore, we show that with SIFTA, we can automatically learn prompts for PrismaDV's modules that outperform prompts written by hand or generated from a generic prompt optimizer. We publicly release our benchmarks and prototype implementation.

cs.LG↗
arXiv2026

Knapsack Optimization-based Schema Linking for LLM-based Text-to-SQL Generation

Generating SQLs from user queries is a long-standing challenge, where the accuracy of initial schema linking significantly impacts subsequent SQL generation performance. However, current schema linking models still struggle with missing relevant schema elements or an excess of redundant ones. A crucial reason for this is that commonly used metrics, recall and precision, fail to capture relevant element missing and thus cannot reflect actual schema linking performance. Motivated by this, we propose enhanced schema linking metrics by introducing a \textbf{restricted missing indicator}. Accordingly, we introduce \textbf{\underline{K}n\underline{a}psack optimization-based \underline{S}chema \underline{L}inking \underline{A}pproach (KaSLA)}, a plug-in schema linking method designed to prevent the missing of relevant schema elements while minimizing the inclusion of redundant ones. KaSLA employs a hierarchical linking strategy that first identifies the optimal table linking and subsequently links columns within the selected table to reduce linking candidate space. In each linking process, it utilizes a knapsack optimization approach to link potentially relevant elements while accounting for a limited tolerance of potentially redundant ones. With this optimization, KaSLA-1.6B achieves superior schema linking results compared to large-scale LLMs, including DeepSeek-V3 with the state-of-the-art (SOTA) schema linking method. Extensive experiments on Spider and BIRD benchmarks verify that KaSLA can significantly improve the SQL generation performance of SOTA Text2SQL models by substituting their schema linking processes. The code is available at https://github.com/DEEP-PolyU/KaSLA.

cs.CL↗
arXiv2026

Matched and Euclidean-Mismatched Decoding on Fourier-Curve Constellations with Tangent Noise

We study matched and Euclidean-mismatched decoding on finite Fourier-curve constellations with tangent-space artificial noise. Each hypothesis induces a Gaussian law with symbol-dependent rank-one covariance. We derive exact Euclidean pairwise errors for arbitrary pairs and an exact Gaussian-expectation representation for matched decoding on bilaterally tangent-orthogonal pairs. For uniform even constellations, the Euclidean side yields explicit distance spectra and symbol-error bounds across all offset classes; the matched side is exact on antipodal pairs and benchmarked numerically at the full-codebook level via Monte Carlo. By isolating the detection-theoretic consequence of tangent-space artificial noise, these results clarify analytically how noise fraction and constellation density enter the mismatch behavior; secrecy-rate implications require additional channel and adversary modeling.

cs.IT↗
arXiv2026

AdaTracker: Learning Adaptive In-Context Policy for Cross-Embodiment Active Visual Tracking

Realizing active visual tracking with a single unified model across diverse robots is challenging, as the physical constraints and motion dynamics vary drastically from one platform to another. Existing approaches typically train separate models for each embodiment, leading to poor scalability and limited generalization. To address this, we propose AdaTracker, an adaptive in-context policy learning framework that robustly tracks targets on diverse robot morphologies. Our key insight is to explicitly model embodiment-specific constraints through an Embodiment Context Encoder, which infers embodiment-specific constraints from history. This contextual representation dynamically modulates a Context-Aware Policy, enabling it to infer optimal control actions for unseen embodiments in a zero-shot manner. To enhance robustness, we introduce two auxiliary objectives to ensure accurate context identification and temporal consistency. Experiments in both simulation and the real world demonstrate that AdaTracker significantly outperforms state-of-the-art methods in cross-embodiment generalization, sample efficiency, and zero-shot adaptation.

cs.RO↗
arXiv2026

Exploring Spatial Intelligence from a Generative Perspective

Spatial intelligence is essential for multimodal large language models, yet current benchmarks largely assess it only from an understanding perspective. We ask whether modern generative or unified multimodal models also possess generative spatial intelligence (GSI), the ability to respect and manipulate 3D spatial constraints during image generation, and whether such capability can be measured or improved. We introduce GSI-Bench, the first benchmark designed to quantify GSI through spatially grounded image editing. It consists of two complementary components: GSI-Real, a high-quality real-world dataset built via a 3D-prior-guided generation and filtering pipeline, and GSI-Syn, a large-scale synthetic benchmark with controllable spatial operations and fully automated labeling. Together with a unified evaluation protocol, GSI-Bench enables scalable, model-agnostic assessment of spatial compliance and editing fidelity. Experiments show that fine-tuning unified multimodal models on GSI-Syn yields substantial gains on both synthetic and real tasks and, strikingly, also improves downstream spatial understanding. This provides the first clear evidence that generative training can tangibly strengthen spatial reasoning, establishing a new pathway for advancing spatial intelligence in multimodal models.

cs.CV↗
arXiv2026

Atomic-scale origin of charge density wave-driven metal-semiconductor transition in an incommensurately modulated metal-organic framework

The intrinsic incommensurate charge density wave in metal-organic frameworks has remained elusive due to the lack of direct evidence linking atomic-scale structural modulation to macroscopic electronic properties. Using high-quality Pr3HHTP2 (HHTP = 2,3,6,7,10,11-hexahydroxytriphenylene) single crystals as a model system, we precisely resolve, for the first time, the incommensurately modulated structure of a conductive metal-organic framework at 100 K (modulation vector q = 0.39143(12) c*) via temperature-dependent single-crystal X-ray diffraction. The subsequent observation of a reversible metal-semiconductor transition around 350 K, which perfectly synchronizes with the disappearance of the structural modulation, provides convincing evidence for the electronic origin of the lattice distortion. Guest water molecules stabilize the modulated phase by synergistically regulating the relative rotation of the linkers and the interlayer spacing, thereby optimizing the inter-linker interactions. This work establishes a concrete experimental criterion for one-dimensional charge density wave in metal-organic frameworks and provides an ideal platform for probing coupled electronic-lattice modulations.

cond-mat.mtrl-sci↗
arXiv2026

Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond

Large-scale data collection is essential for developing personalized training data, mitigating the shortage of training data, and fine-tuning specialized models. However, creating high-quality datasets quickly and accurately remains a challenge due to annotation errors, the substantial time and costs associated with human labor. To address these issues, we propose Automatic Dataset Construction (ADC), an innovative methodology that automates dataset creation with negligible cost and high efficiency. Taking the image classification task as a starting point, ADC leverages LLMs for the detailed class design and code generation to collect relevant samples via search engines, significantly reducing the need for manual annotation and speeding up the data generation process. To demonstrate ADC at scale, we construct Clothing-ADC: a dataset of over 1 million images spanning 12 main classes and 12,000 fine-grained subclasses. Our automated curation achieves 79\% agreement with human annotators and reduces label noise from 22.2\% to 10.7\%. Despite these advantages, ADC also encounters real-world challenges such as label errors (label noise) and imbalanced data distributions (label bias). We provide open-source software that incorporates existing methods for label error detection, robust learning under noisy and biased data, ensuring a higher-quality training data and more robust model training procedure. Furthermore, we design three benchmark datasets focused on label noise detection, label noise learning, and class-imbalanced learning. These datasets are vital because there are few existing datasets specifically for label noise detection, despite its importance. Finally, we evaluate the performance of existing popular methods on these datasets, thereby facilitating further research in the field.

cs.AI↗
arXiv2026

STCast: Adaptive Boundary Alignment for Global and Regional Weather Forecasting

To gain finer regional forecasts, many works have explored the regional integration from the global atmosphere, e.g., by solving boundary equations in physics-based methods or cropping regions from global forecasts in data-driven methods. However, the effectiveness of these methods is often constrained by static and imprecise regional boundaries, resulting in poor generalization ability. To address this issue, we propose Spatial-Temporal Weather Forecasting (STCast), a novel AI-driven framework for adaptive regional boundary optimization and dynamic monthly forecast allocation. Specifically, our approach employs a Spatial-Aligned Attention (SAA) mechanism, which aligns global and regional spatial distributions to initialize boundaries and adaptively refines them based on attention-derived alignment patterns. Furthermore, we design a Temporal Mixture-of-Experts (TMoE) module, where atmospheric variables from distinct months are dynamically routed to specialized experts using a discrete Gaussian distribution, enhancing the model's ability to capture temporal patterns. Beyond global and regional forecasting, we evaluate our STCast on extreme event prediction and ensemble forecasting. Experimental results demonstrate consistent superiority over state-of-the-art methods across all four tasks. Code: https://github.com/chenhao-zju/STCast

cs.LG↗
arXiv2026

Modeling Multi-Dimensional Cognitive States in Large Language Models under Cognitive Crowding

Modeling human cognitive states is essential for advanced artificial intelligence. Existing Large Language Models (LLMs) mainly address isolated tasks such as emotion analysis or stance detection, and fail to capture interactions among cognitive dimensions defined in psychology, including emotion, thinking style, stance, and intention. To bridge this gap, we construct CognitiveBench, the first benchmark with unified annotations across the above four dimensions. Experiments on CognitiveBench show that although LLMs perform well on single dimension tasks, their performance drops sharply in joint multi-dimensional modeling. Using Gromov $δ$-hyperbolicity analysis, we find that CognitiveBench exhibits a strong hierarchical structure. We attribute the performance bottleneck to ``Cognitive Crowding'', where hierarchical cognitive states require exponential representational space, while the Euclidean space of LLMs grows only polynomially, causing representation overlap and degraded performance. To address this mismatch, we propose HyCoLLM, which models cognitive states in hyperbolic space and aligns LLM representations via Hyperbolic Guided Alignment Tuning. Results show that HyCoLLM substantially improves multi-dimensional cognitive understanding, allowing 8B parameter model to outperform strong baselines, including GPT-4o.

cs.CL↗