Searcharxiv⌕ Search

arXiv subjects

Jiajia Chen

Publications and source records attributed to Jiajia Chen.

At least 19 recordsLinked to original sources
arXiv2026

RhinoVLA Technical Report

Vision-Language-Action (VLA) models have shown strong potential for robotic manipulation, but real-time deployment on edge hardware remains challenging. In this work, we identify VLM visual and context tokens as a major source of deployment latency: for GEMM-dominated projection operators, computation grows linearly with the number of input tokens when model dimensions are fixed. Motivated by this observation, we propose RhinoVLA, a deployment-oriented VLA model co-designed with the Huixi R1 edge SoC. RhinoVLA adopts a token-efficient Qwen3-VL backbone and a continuous Action Expert, reducing the VLM-side token and computation burden while preserving pretrained multimodal capability. To support cross-robot learning, RhinoVLA further introduces a unified interface that combines View Registry, 72D physical state-action slot space, and robotinstance LoRA, allowing heterogeneous robot observations and action schemas to be aligned under a shared policy. On the deployment side, RhinoVLA is optimized through hardware-aware compilation, mixed-precision execution, and parallel visual encoding. Experiments show that RhinoVLA achieves downstream performance comparable to π0.5 at a similar parameter scale, while reaching 11.69 Hz end-to-end inference on Huixi R1, meeting the 10 Hz real-time closedloop control target. The project will be open-sourced at https://github.com/HuixiAI/RhinoVLA.

cs.RO↗
arXiv2026

RSTNet: Enhancing Small-Target Recognition in Noisy SAR Imagery via Robust Feature Learning and Distribution-Aware Regression

SAR supports all-day-and-night oceanic observation, yet vessel identification from SAR images is hampered by speckle noise, intricate land-sea backgrounds and dim miniature vessels, yielding numerous false identifications and missed targets. We develop an SAR-adaptive stable detection model RSTNet based on YOLOv8. A large-kernel channel-separated denoising unit eliminates noise and reserves delicate vessel features; parallel patch-aware attention enhances multi-scale feature extraction for miniature objects; NWD loss substitutes conventional IoU loss to achieve accurate bounding box regression. The proposed model outperforms the original YOLOv8 on the SSDD dataset with 97.0% precision, 95.1% recall and 98.9% mAP@0.5. Validations on the HRSID dataset verify its favorable generalization capacity for coastal miniature vessels. Therefore, our work delivers an effective technical scheme for ocean observation imaging with noisy miniature targets. The source code is available at https://github.com/renhcmhx/SAR.git.

cs.CV↗
arXiv2025

MindRec: A Diffusion-driven Coarse-to-Fine Paradigm for Generative Recommendation

Recent advancements in large language model-based recommendation systems often represent items as text or semantic IDs and generate recommendations in an auto-regressive manner. However, due to the left-to-right greedy decoding strategy and the unidirectional logical flow, such methods often fail to produce globally optimal recommendations. In contrast, human reasoning does not follow a rigid left-to-right sequence. Instead, it often begins with keywords or intuitive insights, which are then refined and expanded. Inspired by this fact, we propose MindRec, a diffusion-driven coarse-to-fine generative paradigm that emulates human thought processes. Built upon a diffusion language model, MindRec departs from auto-regressive generation by leveraging a masked diffusion process to reconstruct items in a flexible, non-sequential manner. Particularly, our method first generates key tokens that reflect user preferences, and then expands them into the complete item, enabling adaptive and human-like generation. To further emulate the structured nature of human decision-making, we organize items into a hierarchical category tree. This structure guides the model to first produce the coarse-grained category and then progressively refine its selection through finer-grained subcategories before generating the specific item. To mitigate the local optimum problem inherent in greedy decoding, we design a novel beam search algorithm, Diffusion Beam Search, tailored for our mind-inspired generation paradigm. Experimental results demonstrate that MindRec yields a 9.5\% average improvement in top-1 accuracy over state-of-the-art methods, highlighting its potential to enhance recommendation performance. The implementation is available via https://github.com/Mr-Peach0301/MindRec.

cs.IR↗
arXiv2025

Wurtzite AlScN/AlN Superlattice Ferroelectrics Enable Endurance Beyond 1010 Cycles

Wurtzite ferroelectrics are rapidly emerging as a promising material class for next-generation non-volatile memory technologies, owing to their large remanent polarization, intrinsically ordered three-dimensional crystal structure, and full compatibility with CMOS processes and back-end-of-line (BEOL) integration. However, their practical implementation remains critically constrained by a severe endurance bottleneck: under conditions where the remanent polarization (2Pr) reaches or exceeds 200 uC/cm^2, devices typically undergo catastrophic failure before reaching 10^8 cycles. Here, we report a vacancy-confining superlattice strategy that addresses this limitation, achieving reliable ferroelectric switching beyond 10^10 cycles while preserving saturated polarization (2Pr >= 200 uC/cm^2). This is achieved by embedding periodic ultrathin AlN layers within AlScN films, forming wurtzite AlScN/AlN superlattices, in conjunction with a dynamic recovery protocol that actively stabilizes the defect landscape throughout repeated cycling. Atomic-resolution imaging and EELS spectrum imaging technique, supported by first-principles calculations, reveal a self-regulated defect topology in which nitrogen vacancies are spatially confined by heterostructure energy barriers and dynamically re-trapped into energetically favorable lattice sites. This dual spatial-energetic confinement mechanism effectively inhibits both long-range percolative migration and local defect clustering, enabling such an ultrahigh endurance exceeding 10^10 cycles and limiting polarization degradation to below 3% after 10^9 cycles. These findings establish nitrogen vacancy topology stabilization as a foundational design principle for reliable operation of wurtzite ferroelectrics, providing a scalable and CMOS-compatible platform for future high-endurance ferroelectric memory technologies.

cond-mat.mtrl-sci↗
arXiv2024

Position-aware Graph Transformer for Recommendation

Collaborative recommendation fundamentally involves learning high-quality user and item representations from interaction data. Recently, graph convolution networks (GCNs) have advanced the field by utilizing high-order connectivity patterns in interaction graphs, as evidenced by state-of-the-art methods like PinSage and LightGCN. However, one key limitation has not been well addressed in existing solutions: capturing long-range collaborative filtering signals, which are crucial for modeling user preference. In this work, we propose a new graph transformer (GT) framework -- \textit{Position-aware Graph Transformer for Recommendation} (PGTR), which combines the global modeling capability of Transformer blocks with the local neighborhood feature extraction of GCNs. The key insight is to explicitly incorporate node position and structure information from the user-item interaction graph into GT architecture via several purpose-designed positional encodings. The long-range collaborative signals from the Transformer block are then combined linearly with the local neighborhood features from the GCN backbone to enhance node embeddings for final recommendations. Empirical studies demonstrate the effectiveness of the proposed PGTR method when implemented on various GCN-based backbones across four real-world datasets, and the robustness against interaction sparsity as well as noise.

cs.IR↗
arXiv2023

How Graph Convolutions Amplify Popularity Bias for Recommendation?

Graph convolutional networks (GCNs) have become prevalent in recommender system (RS) due to their superiority in modeling collaborative patterns. Although improving the overall accuracy, GCNs unfortunately amplify popularity bias -- tail items are less likely to be recommended. This effect prevents the GCN-based RS from making precise and fair recommendations, decreasing the effectiveness of recommender systems in the long run. In this paper, we investigate how graph convolutions amplify the popularity bias in RS. Through theoretical analyses, we identify two fundamental factors: (1) with graph convolution (\textit{i.e.,} neighborhood aggregation), popular items exert larger influence than tail items on neighbor users, making the users move towards popular items in the representation space; (2) after multiple times of graph convolution, popular items would affect more high-order neighbors and become more influential. The two points make popular items get closer to almost users and thus being recommended more frequently. To rectify this, we propose to estimate the amplified effect of popular nodes on each node's representation, and intervene the effect after each graph convolution. Specifically, we adopt clustering to discover highly-influential nodes and estimate the amplification effect of each node, then remove the effect from the node embeddings at each graph convolution layer. Our method is simple and generic -- it can be used in the inference stage to correct existing models rather than training a new model from scratch, and can be applied to various GCN models. We demonstrate our method on two representative GCN backbones LightGCN and UltraGCN, verifying its ability in improving the recommendations of tail items without sacrificing the performance of popular items. Codes are open-sourced \footnote{https://github.com/MEICRS/DAP}.

cs.IR↗
arXiv2023

Optical Networks and Interconnects

The rapid evolution of communication technologies such as 5G and beyond, rely on optical networks to support the challenging and ambitious requirements that include both capacity and reliability. This chapter begins by giving an overview of the evolution of optical access networks, focusing on Passive Optical Networks (PONs). The development of the different PON standards and requirements aiming at longer reach, higher client count and delivered bandwidth are presented. PON virtualization is also introduced as the flexibility enabler. Triggered by the increase of bandwidth supported by access and aggregation network segments, core networks have also evolved, as presented in the second part of the chapter. Scaling the physical infrastructure requires high investment and hence, operators are considering alternatives to optimize the use of the existing capacity. This chapter introduces different planning problems such as Routing and Spectrum Assignment problems, placement problems for regenerators and wavelength converters, and how to offer resilience to different failures. An overview of control and management is also provided. Moreover, motivated by the increasing importance of data storage and data processing, this chapter also addresses different aspects of optical data center interconnects. Data centers have become critical infrastructure to operate any service. They are also forced to take advantage of optical technology in order to keep up with the growing capacity demand and power consumption. This chapter gives an overview of different optical data center network architectures as well as some expected directions to improve the resource utilization and increase the network capacity.

cs.NI↗
arXiv2022

GDSRec: Graph-Based Decentralized Collaborative Filtering for Social Recommendation

Generating recommendations based on user-item interactions and user-user social relations is a common use case in web-based systems. These connections can be naturally represented as graph-structured data and thus utilizing graph neural networks (GNNs) for social recommendation has become a promising research direction. However, existing graph-based methods fails to consider the bias offsets of users (items). For example, a low rating from a fastidious user may not imply a negative attitude toward this item because the user tends to assign low ratings in common cases. Such statistics should be considered into the graph modeling procedure. While some past work considers the biases, we argue that these proposed methods only treat them as scalars and can not capture the complete bias information hidden in data. Besides, social connections between users should also be differentiable so that users with similar item preference would have more influence on each other. To this end, we propose Graph-Based Decentralized Collaborative Filtering for Social Recommendation (GDSRec). GDSRec treats the biases as vectors and fuses them into the process of learning user and item representations. The statistical bias offsets are captured by decentralized neighborhood aggregation while the social connection strength is defined according to the preference similarity and then incorporated into the model design. We conduct extensive experiments on two benchmark datasets to verify the effectiveness of the proposed model. Experimental results show that the proposed GDSRec achieves superior performance compared with state-of-the-art related baselines. Our implementations are available in \url{https://github.com/MEICRS/GDSRec}.

cs.IR↗
arXiv2021

Intelligent Online Selling Point Extraction for E-Commerce Recommendation

In the past decade, automatic product description generation for e-commerce have witnessed significant advancement. As the services provided by e-commerce platforms become diverse, it is necessary to dynamically adapt the patterns of descriptions generated. The selling point of products is an important type of product description for which the length should be as short as possible while still conveying key information. In addition, this kind of product description should be eye-catching to the readers. Currently, product selling points are normally written by human experts. Thus, the creation and maintenance of these contents incur high costs. These costs can be significantly reduced if product selling points can be automatically generated by machines. In this paper, we report our experience developing and deploying the Intelligent Online Selling Point Extraction (IOSPE) system to serve the recommendation system in the JD.com e-commerce platform. Since July 2020, IOSPE has become a core service for 62 key categories of products (covering more than 4 million products). So far, it has generated more than 0.1 billion selling points, thereby significantly scaling up the selling point creation operation and saving human labour. These IOSPE generated selling points have increased the click-through rate (CTR) by 1.89\% and the average duration the customers spent on the products by more than 2.03\% compared to the previous practice, which are significant improvements for such a large-scale e-commerce platform.

cs.IR↗
arXiv2021

Automatic Product Copywriting for E-Commerce

Product copywriting is a critical component of e-commerce recommendation platforms. It aims to attract users' interest and improve user experience by highlighting product characteristics with textual descriptions. In this paper, we report our experience deploying the proposed Automatic Product Copywriting Generation (APCG) system into the JD.com e-commerce product recommendation platform. It consists of two main components: 1) natural language generation, which is built from a transformer-pointer network and a pre-trained sequence-to-sequence model based on millions of training data from our in-house platform; and 2) copywriting quality control, which is based on both automatic evaluation and human screening. For selected domains, the models are trained and updated daily with the updated training data. In addition, the model is also used as a real-time writing assistant tool on our live broadcast platform. The APCG system has been deployed in JD.com since Feb 2021. By Sep 2021, it has generated 2.53 million product descriptions, and improved the overall averaged click-through rate (CTR) and the Conversion Rate (CVR) by 4.22% and 3.61%, compared to baselines, respectively on a year-on-year basis. The accumulated Gross Merchandise Volume (GMV) made by our system is improved by 213.42%, compared to the number in Feb 2021.

cs.CL↗
arXiv2020

Scalable Federated Learning over Passive Optical Networks

Two-step aggregation is introduced to facilitate scalable federated learning (SFL) over passive optical networks (PONs). Results reveal that the SFL keeps the required PON upstream bandwidth constant regardless of the number of involved clients, while bringing ~10% learning accuracy improvement.

cs.NI↗
arXiv2019

Bandwidth Slicing to Boost Federated Learning in Edge Computing

Bandwidth slicing is introduced to support federated learning in edge computing to assure low communication delay for training traffic. Results reveal that bandwidth slicing significantly improves training efficiency while achieving good learning accuracy.

cs.NI↗
arXiv2018

High-speed PAM4-based Optical SDM Interconnects with Directly Modulated Long-wavelength VCSEL

This paper reports the demonstration of high-speed PAM-4 transmission using a 1.5-μm single-mode vertical cavity surface emitting laser (SM-VCSEL) over multicore fiber with 7 cores over different distances. We have successfully generated up to 70 Gbaud 4-level pulse amplitude modulation (PAM-4) signals with a VCSEL in optical back-to-back, and transmitted 50 Gbaud PAM-4 signals over both 1-km dispersion-uncompensated and 10-km dispersion-compensated in each core, enabling a total data throughput of 700 Gbps over the 7-core fiber. Moreover, 56 Gbaud PAM-4 over 1-km has also been shown, whereby unfortunately not all cores provide the required 3.8 $\times$ 10 $^{-3}$ bit error rate (BER) for the 7% overhead-hard decision forward error correction (7% OH HDFEC). The limited bandwidth of the VCSEL and the adverse chromatic dispersion of the fiber are suppressed with pre-equalization based on accurate end-to-end channel characterizations. With a digital post-equalization, BER performance below the 7% OH-HDFEC limit is achieved over all cores. The demonstrated results show a great potential to realize high-capacity and compact short-reach optical interconnects for data centers.

eess.SP↗
arXiv2018

140 Gbaud On-Off Keying Links in C-Band for Short-Reach Optical Interconnects

We demonstrate 140 Gbaud intensity modulated direct detection dispersion-uncompensated links with Mach Zehnder modulator and distributed feedback travelling-wave electro-absorption modulator over 5500 and 960 meters of standard single mode fibre, respectively, enabled by compact packaged ultra-high speed InP-based 2:1-Selector.

eess.SP↗