arXiv subjects

Hao Wang

Publications and source records attributed to Hao Wang.

At least 757 records · Page 42Linked to original sources

arXiv2025

Operational Semantics for Crystality: A Smart Contract Language for Parallel EVMs

The increasing demand for scalable blockchain has driven research into parallel execution models for smart contracts. Crystality is a novel smart contract programming language designed for parallel Ethereum Virtual Machines (EVMs), enabling fine-grained concurrency through Programmable Contract Scopes and Asynchronous Functional Relay. This paper presents the first formal structural operational semantics for Crystality, providing a rigorous framework to reason about its execution. We mechanize the syntax and semantics of Crystality in the theorem-proving assistant Coq, enabling formal verification of correctness properties. As a case study, we verify a simplified token transfer function, demonstrating the applicability of our semantics in ensuring smart contract correctness. Our work lays the foundation for formally verified parallel smart contracts, contributing to the security and scalability of blockchain systems.

cs.PL↗

arXiv2025

An Empirical Study on Prompt Compression for Large Language Models

Prompt engineering enables Large Language Models (LLMs) to perform a variety of tasks. However, lengthy prompts significantly increase computational complexity and economic costs. To address this issue, we study six prompt compression methods for LLMs, aiming to reduce prompt length while maintaining LLM response quality. In this paper, we present a comprehensive analysis covering aspects such as generation performance, model hallucinations, efficacy in multimodal tasks, word omission analysis, and more. We evaluate these methods across 13 datasets, including news, scientific articles, commonsense QA, math QA, long-context QA, and VQA datasets. Our experiments reveal that prompt compression has a greater impact on LLM performance in long contexts compared to short ones. In the Longbench evaluation, moderate compression even enhances LLM performance. Our code and data is available at https://github.com/3DAgentWorld/Toolkit-for-Prompt-Compression.

cs.CL↗

arXiv2025

Killing Two Birds with One Stone: Unifying Retrieval and Ranking with a Single Generative Recommendation Model

In recommendation systems, the traditional multi-stage paradigm, which includes retrieval and ranking, often suffers from information loss between stages and diminishes performance. Recent advances in generative models, inspired by natural language processing, suggest the potential for unifying these stages to mitigate such loss. This paper presents the Unified Generative Recommendation Framework (UniGRF), a novel approach that integrates retrieval and ranking into a single generative model. By treating both stages as sequence generation tasks, UniGRF enables sufficient information sharing without additional computational costs, while remaining model-agnostic. To enhance inter-stage collaboration, UniGRF introduces a ranking-driven enhancer module that leverages the precision of the ranking stage to refine retrieval processes, creating an enhancement loop. Besides, a gradient-guided adaptive weighter is incorporated to dynamically balance the optimization of retrieval and ranking, ensuring synchronized performance improvements. Extensive experiments demonstrate that UniGRF significantly outperforms existing models on benchmark datasets, confirming its effectiveness in facilitating information transfer. Ablation studies and further experiments reveal that UniGRF not only promotes efficient collaboration between stages but also achieves synchronized optimization. UniGRF provides an effective, scalable, and compatible framework for generative recommendation systems.

cs.IR↗

arXiv2025

ManipDreamer: Boosting Robotic Manipulation World Model with Action Tree and Visual Guidance

While recent advancements in robotic manipulation video synthesis have shown promise, significant challenges persist in ensuring effective instruction-following and achieving high visual quality. Recent methods, like RoboDreamer, utilize linguistic decomposition to divide instructions into separate lower-level primitives, conditioning the world model on these primitives to achieve compositional instruction-following. However, these separate primitives do not consider the relationships that exist between them. Furthermore, recent methods neglect valuable visual guidance, including depth and semantic guidance, both crucial for enhancing visual quality. This paper introduces ManipDreamer, an advanced world model based on the action tree and visual guidance. To better learn the relationships between instruction primitives, we represent the instruction as the action tree and assign embeddings to tree nodes, each instruction can acquire its embeddings by navigating through the action tree. The instruction embeddings can be used to guide the world model. To enhance visual quality, we combine depth and semantic guidance by introducing a visual guidance adapter compatible with the world model. This visual adapter enhances both the temporal and physical consistency of video generation. Based on the action tree and visual guidance, ManipDreamer significantly boosts the instruction-following ability and visual quality. Comprehensive evaluations on robotic manipulation benchmarks reveal that ManipDreamer achieves large improvements in video quality metrics in both seen and unseen tasks, with PSNR improved from 19.55 to 21.05, SSIM improved from 0.7474 to 0.7982 and reduced Flow Error from 3.506 to 3.201 in unseen tasks, compared to the recent RoboDreamer model. Additionally, our method increases the success rate of robotic manipulation tasks by 2.5% in 6 RLbench tasks on average.

cs.RO↗

arXiv2025

EG-Gaussian: Epipolar Geometry and Graph Network Enhanced 3D Gaussian Splatting

In this paper, we explore an open research problem concerning the reconstruction of 3D scenes from images. Recent methods have adopt 3D Gaussian Splatting (3DGS) to produce 3D scenes due to its efficient training process. However, these methodologies may generate incomplete 3D scenes or blurred multiviews. This is because of (1) inaccurate 3DGS point initialization and (2) the tendency of 3DGS to flatten 3D Gaussians with the sparse-view input. To address these issues, we propose a novel framework EG-Gaussian, which utilizes epipolar geometry and graph networks for 3D scene reconstruction. Initially, we integrate epipolar geometry into the 3DGS initialization phase to enhance initial 3DGS point construction. Then, we specifically design a graph learning module to refine 3DGS spatial features, in which we incorporate both spatial coordinates and angular relationships among neighboring points. Experiments on indoor and outdoor benchmark datasets demonstrate that our approach significantly improves reconstruction accuracy compared to 3DGS-based methods.

cs.CV↗

arXiv2025

LearnAct: Few-Shot Mobile GUI Agent with a Unified Demonstration Benchmark

Mobile GUI agents show promise in automating tasks but face generalization challenges in diverse real-world scenarios. Traditional approaches using pre-training or fine-tuning with massive datasets struggle with the diversity of mobile applications and user-specific tasks. We propose enhancing mobile GUI agent capabilities through human demonstrations, focusing on improving performance in unseen scenarios rather than pursuing universal generalization through larger datasets. To realize this paradigm, we introduce LearnGUI, the first comprehensive dataset specifically designed for studying demonstration-based learning in mobile GUI agents, comprising 2,252 offline tasks and 101 online tasks with high-quality human demonstrations. We further develop LearnAct, a sophisticated multi-agent framework that automatically extracts knowledge from demonstrations to enhance task completion. This framework integrates three specialized agents: DemoParser for knowledge extraction, KnowSeeker for relevant knowledge retrieval, and ActExecutor for demonstration-enhanced task execution. Our experimental results show significant performance gains in both offline and online evaluations. In offline assessments, a single demonstration improves model performance, increasing Gemini-1.5-Pro's accuracy from 19.3% to 51.7%. In online evaluations, our framework enhances UI-TARS-7B-SFT's task success rate from 18.1% to 32.8%. LearnAct framework and LearnGUI benchmark establish demonstration-based learning as a promising direction for more adaptable, personalized, and deployable mobile GUI agents.

cs.HC↗

arXiv2025

MoS2 Based 2D Material Photodetector Array with high Pixel Density

Arrays of photodetector-based pixel sensors are ubiquitous in modern devices, such as smart phone cameras, automobiles, drones, laptops etc. Two dimensional (2D) material-based photodetector arrays are a relevant candidate, especially for applications demanding planar formfactors. However, shortcomings in pixel density and prototyping without cross contamination limit technology adoption and impact. Also, while 2D material detectors offer high absorption, graphene's closed bandgap results in undesirably high dark currents. Here, we introduce the experimental demonstration of dense planar photodetector arrays. We demonstrate a micrometer narrow pitched 2D detector pixels and show this approach's repeatability by verifying performance of a 16-pixel array. Such dense and repeatable detector realization is enabled by a novel, selective, contamination free 2D material transfer system, that we report here in automated operation. The so realized photodetectors responsivity peaks at 0.8 A/W. Furthermore, we achieve uniform detector performance via bias voltage tuning calibration to maximize deployment. Finally, we demonstrate 2D arrayed photodetectors not only on a silicon chip platform but also demonstrate and very array performance on flexible polymer substrates. Densely arrayed, flat, bendable, and uniform performing photodetector pixels enable emerging technologies in the space where lightweight and reliable performance is required, such as for smart phones and emerging AR/VR markets, but smart gadgets, wearables, and for SWAP constrained aviation and space platforms.

physics.optics↗

arXiv2025

Abnormal Mutations: Evolution Strategies Don't Require Gaussianity

The mutation process in evolution strategies has been interlinked with the normal distribution since its inception. Many lines of reasoning have been given for this strong dependency, ranging from maximum entropy arguments to the need for isotropy. However, some theoretical results suggest that other distributions might lead to similar local convergence properties. This paper empirically shows that a wide range of evolutionary strategies, from the (1+1)-ES to CMA-ES, show comparable optimization performance when using a mutation distribution other than the standard Gaussian. Replacing it with, e.g., uniformly distributed mutations, does not deteriorate the performance of ES, when using the default adaptation mechanism for the strategy parameters. We observe that these results hold not only for the sphere model but also for a wider range of benchmark problems.

cs.NE↗

arXiv2025

HGFormer: Topology-Aware Vision Transformer with HyperGraph Learning

The computer vision community has witnessed an extensive exploration of vision transformers in the past two years. Drawing inspiration from traditional schemes, numerous works focus on introducing vision-specific inductive biases. However, the implicit modeling of permutation invariance and fully-connected interaction with individual tokens disrupts the regional context and spatial topology, further hindering higher-order modeling. This deviates from the principle of perceptual organization that emphasizes the local groups and overall topology of visual elements. Thus, we introduce the concept of hypergraph for perceptual exploration. Specifically, we propose a topology-aware vision transformer called HyperGraph Transformer (HGFormer). Firstly, we present a Center Sampling K-Nearest Neighbors (CS-KNN) algorithm for semantic guidance during hypergraph construction. Secondly, we present a topology-aware HyperGraph Attention (HGA) mechanism that integrates hypergraph topology as perceptual indications to guide the aggregation of global and unbiased information during hypergraph messaging. Using HGFormer as visual backbone, we develop an effective and unitive representation, achieving distinct and detailed scene depictions. Empirical experiments show that the proposed HGFormer achieves competitive performance compared to the recent SoTA counterparts on various visual benchmarks. Extensive ablation and visualization studies provide comprehensive explanations of our ideas and contributions.

cs.CV↗

arXiv2025

Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learning

Continual learning in large language models (LLMs) is prone to catastrophic forgetting, where adapting to new tasks significantly degrades performance on previously learned ones. Existing methods typically rely on low-rank, parameter-efficient updates that limit the model's expressivity and introduce additional parameters per task, leading to scalability issues. To address these limitations, we propose a novel continual full fine-tuning approach leveraging adaptive singular value decomposition (SVD). Our method dynamically identifies task-specific low-rank parameter subspaces and constrains updates to be orthogonal to critical directions associated with prior tasks, thus effectively minimizing interference without additional parameter overhead or storing previous task gradients. We evaluate our approach extensively on standard continual learning benchmarks using both encoder-decoder (T5-Large) and decoder-only (LLaMA-2 7B) models, spanning diverse tasks including classification, generation, and reasoning. Empirically, our method achieves state-of-the-art results, up to 7% higher average accuracy than recent baselines like O-LoRA, and notably maintains the model's general linguistic capabilities, instruction-following accuracy, and safety throughout the continual learning process by reducing forgetting to near-negligible levels. Our adaptive SVD framework effectively balances model plasticity and knowledge retention, providing a practical, theoretically grounded, and computationally scalable solution for continual learning scenarios in large language models.

cs.LG↗

arXiv2025

Note on the Universality of Parameterized IQP Circuits with Hidden Units for Generating Probability Distributions

In a series of recent works, an interesting quantum generative model based on parameterized instantaneous polynomial quantum (IQP) circuits has emerged as they can be trained efficiently classically using any loss function that depends only on the expectation values of observables of the model. The model is proven not to be universal for generating arbitrary distributions, but it is suspected that marginals can be - much like Boltzmann machines achieve universality by utilizing hidden (traced-out in quantum jargon) layers. In this short note, we provide two simple proofs of this fact. The first is near-trivial and asymptotic, and the second shows universality can be achieved with a reasonable number of additional qubits.

quant-ph↗

arXiv2025

High Gain and Broadband Metalens Antenna for Terahertz Communication

Terahertz (THz) metalens antennas with compact planar structures have demonstrated significant potential in enhancing gain and aperture efficiency through beam convergence. However, research on THz wireless communication systems utilizing metalens antennas remains limited, primarily due to insufficient collaborative enhancement in gain and bandwidth in THz transceiver design. In this paper, we propose a high gain metalens antenna transceiver and demonstrate its application for THz communication. The system employs a horn antenna integrated with a 3D-printed bracket to enhance the metalens gain and operating bandwidth, where the metalens adopts a "sandwich" architecture composed of a V-shaped copper resonator, a dielectric substrate, and a grating. The resonant design inside the metalens facilitates high polarization conversion efficiency and full phase modulation across a 0° to 360° range at frequency between 0.20 to 0.30 THz band. Experimental results demonstrate a peak gain of 36.1 dBi and aperture efficiency of 54.45% at 0.244 THz, with a 3 dB bandwidth exceeding 33 GHz. A prototype communication system incorporating the metalens transceiver achieves a bit error rate (BER) reduction by three orders of magnitude compared to conventional horn antennas and supports a maximum data rate of 100 Gbps. This proposed metalens offer a high-gain, compact solution for achieving high data rate THz communications, driving advancements in 6G communication network.

physics.app-ph↗

arXiv2025

Towards Domain Adaptive Neural Contextual Bandits

Contextual bandit algorithms are essential for solving real-world decision making problems. In practice, collecting a contextual bandit's feedback from different domains may involve different costs. For example, measuring drug reaction from mice (as a source domain) and humans (as a target domain). Unfortunately, adapting a contextual bandit algorithm from a source domain to a target domain with distribution shift still remains a major challenge and largely unexplored. In this paper, we introduce the first general domain adaptation method for contextual bandits. Our approach learns a bandit model for the target domain by collecting feedback from the source domain. Our theoretical analysis shows that our algorithm maintains a sub-linear regret bound even adapting across domains. Empirical results show that our approach outperforms the state-of-the-art contextual bandit algorithms on real-world datasets.

cs.LG↗

arXiv2025

FlowKV: A Disaggregated Inference Framework with Low-Latency KV Cache Transfer and Load-Aware Scheduling

Disaggregated inference has become an essential framework that separates the prefill (P) and decode (D) stages in large language model inference to improve throughput. However, the KV cache transfer faces significant delays between prefill and decode nodes. The block-wise calling method and discontinuous KV cache memory allocation increase the number of calls to the transmission kernel. Additionally, existing frameworks often fix the roles of P and D nodes, leading to computational imbalances. In this paper, we propose FlowKV, a novel disaggregated inference framework, which reduces the average transmission latency of KV cache by 96%, from 0.944s to 0.053s, almost eliminating the transfer time relative to the total request latency by optimizing the KV cache transfer. FlowKV introduces the Load-Aware Scheduler for balanced request scheduling and flexible PD node allocation. This design maximizes hardware resource utilization, achieving peak system throughput across various scenarios, including normal, computational imbalance, and extreme overload conditions. Experimental results demonstrate that FlowKV significantly accelerates inference by 15.2%-48.9% on LongBench dataset compared to the baseline and supports applications with heterogeneous GPUs.

cs.DC↗

arXiv2025

Enhanced Cross-modal 3D Retrieval via Tri-modal Reconstruction

Cross-modal 3D retrieval is a critical yet challenging task, aiming to achieve bi-directional retrieval between 3D and text modalities. Current methods predominantly rely on a certain 3D representation (e.g., point cloud), with few exploiting the 2D-3D consistency and complementary relationships, which constrains their performance. To bridge this gap, we propose to adopt multi-view images and point clouds to jointly represent 3D shapes, facilitating tri-modal alignment (i.e., image, point, text) for enhanced cross-modal 3D retrieval. Notably, we introduce tri-modal reconstruction to improve the generalization ability of encoders. Given point features, we reconstruct image features under the guidance of text features, and vice versa. With well-aligned point cloud and multi-view image features, we aggregate them as multimodal embeddings through fine-grained 2D-3D fusion to enhance geometric and semantic understanding. Recognizing the significant noise in current datasets where many 3D shapes and texts share similar semantics, we employ hard negative contrastive training to emphasize harder negatives with greater significance, leading to robust discriminative embeddings. Extensive experiments on the Text2Shape dataset demonstrate that our method significantly outperforms previous state-of-the-art methods in both shape-to-text and text-to-shape retrieval tasks by a substantial margin.

cs.CV↗

arXiv2025

Alternating Iteratively Reweighted $\ell_1$ and Subspace Newton Algorithms for Nonconvex Sparse Optimization

This paper presents a novel hybrid algorithm for minimizing the sum of a continuously differentiable loss function and a nonsmooth, possibly nonconvex, sparse regularization function. The proposed method alternates between solving a reweighted $\ell_1$-regularized subproblem and performing an inexact subspace Newton step. The reweighted $\ell_1$-subproblem allows for efficient closed-form solutions via the soft-thresholding operator, avoiding the computational overhead of proximity operator calculations. As the algorithm approaches an optimal solution, it maintains a stable support set, ensuring that nonzero components stay uniformly bounded away from zero. It then switches to a perturbed regularized Newton method, further accelerating the convergence. We prove global convergence to a critical point and, under suitable conditions, demonstrate that the algorithm exhibits local linear and quadratic convergence rates. Numerical experiments show that our algorithm outperforms existing methods in both efficiency and solution quality across various model prediction problems.

math.OC↗

arXiv2025

GenVP: Generating Visual Puzzles with Contrastive Hierarchical VAEs

Raven's Progressive Matrices (RPMs) is an established benchmark to examine the ability to perform high-level abstract visual reasoning (AVR). Despite the current success of algorithms that solve this task, humans can generalize beyond a given puzzle and create new puzzles given a set of rules, whereas machines remain locked in solving a fixed puzzle from a curated choice list. We propose Generative Visual Puzzles (GenVP), a framework to model the entire RPM generation process, a substantially more challenging task. Our model's capability spans from generating multiple solutions for one specific problem prompt to creating complete new puzzles out of the desired set of rules. Experiments on five different datasets indicate that GenVP achieves state-of-the-art (SOTA) performance both in puzzle-solving accuracy and out-of-distribution (OOD) generalization in 22 OOD scenarios. Compared to SOTA generative approaches, which struggle to solve RPMs when the feasible solution space increases, GenVP efficiently generalizes to these challenging setups. Moreover, our model demonstrates the ability to produce a wide range of complete RPMs given a set of abstract rules by effectively capturing the relationships between abstract rules and visual object properties.

cs.AI↗

arXiv2025

FTS: A Framework to Find a Faithful TimeSieve

The field of time series forecasting has garnered significant attention in recent years, prompting the development of advanced models like TimeSieve, which demonstrates impressive performance. However, an analysis reveals certain unfaithfulness issues, including high sensitivity to random seeds and minute input noise perturbations. Recognizing these challenges, we embark on a quest to define the concept of \textbf{\underline{F}aithful \underline{T}ime\underline{S}ieve \underline{(FTS)}}, a model that consistently delivers reliable and robust predictions. To address these issues, we propose a novel framework aimed at identifying and rectifying unfaithfulness in TimeSieve. Our framework is designed to enhance the model's stability and resilience, ensuring that its outputs are less susceptible to the aforementioned factors. Experimentation validates the effectiveness of our proposed framework, demonstrating improved faithfulness in the model's behavior. Looking forward, we plan to expand our experimental scope to further validate and optimize our algorithm, ensuring comprehensive faithfulness across a wide range of scenarios. Ultimately, we aspire to make this framework can be applied to enhance the faithfulness of not just TimeSieve but also other state-of-the-art temporal methods, thereby contributing to the reliability and robustness of temporal modeling as a whole.

cs.LG↗