arXiv subjects

Wei Ding

Publications and source records attributed to Wei Ding.

At least 37 records · Page 2Linked to original sources

arXiv2025

SIEVE: Effective Filtered Vector Search with Collection of Indexes

Many real-world tasks such as recommending videos with the kids tag can be reduced to finding most similar vectors associated with hard predicates. This task, filtered vector search, is challenging as prior state-of-the-art graph-based (unfiltered) similarity search techniques quickly degenerate when hard constraints are considered. That is, effective graph-based filtered similarity search relies on sufficient connectivity for reaching the most similar items within just a few hops. To consider predicates, recent works propose modifying graph traversal to visit only the items that may satisfy predicates. However, they fail to offer the just-a-few-hops property for a wide range of predicates: they must restrict predicates significantly or lose efficiency if only a small fraction of items satisfy predicates. We propose an opposite approach: instead of constraining traversal, we build many indexes each serving different predicate forms. For effective construction, we devise a three-dimensional analytical model capturing relationships among index size, search time, and recall, with which we follow a workload-aware approach to pack as many useful indexes as possible into a collection. At query time, the analytical model is employed yet again to discern the one that offers the fastest search at a given recall. We show superior performance and support on datasets with varying selectivities and forms: our approach achieves up to 8.06x speedup while having as low as 1% build time versus other indexes, with less than 2.15x memory of a standard HNSW graph and modest knowledge of past workloads.

cs.DB↗

arXiv2025

Demonstrating Interoperable Channel State Feedback Compression with Machine Learning

Neural network-based compression and decompression of channel state feedback has been one of the most widely studied applications of machine learning (ML) in wireless networks. Various simulation-based studies have shown that ML-based feedback compression can result in reduced overhead and more accurate channel information. However, to the best of our knowledge, there are no real-life proofs of concepts demonstrating the benefits of ML-based channel feedback compression in a practical setting, where the user equipment (UE) and base station have no access to each others' ML models. In this paper, we present a novel approach for training interoperable compression and decompression ML models in a confidential manner, and demonstrate the accuracy of the ensuing models using prototype UEs and base stations. The performance of the ML-based channel feedback is measured both in terms of the accuracy of the reconstructed channel information and achieved downlink throughput gains when using the channel information for beamforming. The reported measurement results demonstrate that it is possible to develop an accurate ML-based channel feedback link without having to share ML models between device and network vendors. These results pave the way for a practical implementation of ML-based channel feedback in commercial 6G networks.

eess.SP↗

arXiv2025

Qwen2.5-VL Technical Report

We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.

cs.CV↗

arXiv2025

SteLLA: A Structured Grading System Using LLMs with RAG

Large Language Models (LLMs) have shown strong general capabilities in many applications. However, how to make them reliable tools for some specific tasks such as automated short answer grading (ASAG) remains a challenge. We present SteLLA (Structured Grading System Using LLMs with RAG) in which a) Retrieval Augmented Generation (RAG) approach is used to empower LLMs specifically on the ASAG task by extracting structured information from the highly relevant and reliable external knowledge based on the instructor-provided reference answer and rubric, b) an LLM performs a structured and question-answering-based evaluation of student answers to provide analytical grades and feedback. A real-world dataset that contains students' answers in an exam was collected from a college-level Biology course. Experiments show that our proposed system can achieve substantial agreement with the human grader while providing break-down grades and feedback on all the knowledge points examined in the problem. A qualitative and error analysis of the feedback generated by GPT4 shows that GPT4 is good at capturing facts while may be prone to inferring too much implication from the given text in the grading task which provides insights into the usage of LLMs in the ASAG system.

cs.CL↗

arXiv2025

TreeMatch: A Fully Unsupervised WSD System Using Dependency Knowledge on a Specific Domain

Word sense disambiguation (WSD) is one of the main challenges in Computational Linguistics. TreeMatch is a WSD system originally developed using data from SemEval 2007 Task 7 (Coarse-grained English All-words Task) that has been adapted for use in SemEval 2010 Task 17 (All-words Word Sense Disambiguation on a Specific Domain). The system is based on a fully unsupervised method using dependency knowledge drawn from a domain specific knowledge base that was built for this task. When evaluated on the task, the system precision performs above the Most Frequent Selection baseline.

cs.CL↗

arXiv2024

AToM-Bot: Embodied Fulfillment of Unspoken Human Needs with Affective Theory of Mind

We propose AToM-Bot, a novel task generation and execution framework for proactive robot-human interaction, which leverages the human mental and physical state inference capabilities of the Vision Language Model (VLM) prompted by the Affective Theory of Mind (AToM). Without requiring explicit commands by humans, AToM-Bot proactively generates and follows feasible tasks to improve general human well-being. When around humans, AToM-Bot first detects current human needs based on inferred human states and observations of the surrounding environment. It then generates tasks to fulfill these needs, taking into account its embodied constraints. We designed 16 daily life scenarios spanning 4 common scenes and tasked the same visual stimulus to 59 human subjects and our robot. We used the similarity between human open-ended answers and robot output, and the human satisfaction scores to metric robot performance. AToM-Bot received high human evaluations in need detection (6.42/7, 91.7%), embodied solution (6.15/7, 87.8%) and task execution (6.17/7, 88.1%). We show that AToM-Bot excels in generating and executing feasible plans to fulfill unspoken human needs. Videos and code are available at https://affective-tom-bot.github.io.

cs.RO↗

arXiv2024

Four-fold truncated double-nested anti-resonant hollow-core fibers with ultralow loss and ultrahigh mode purity

Hollow-core fibers are inherently multimode, making it crucial to filter out higher-order modes within the shortest possible fiber length for applications such as high speed coherent communications and fiber optic gyroscopes. However, current HCF designs face the challenges of simultaneously achieving ultralow fundamental mode loss and ultrahigh HOM suppression. In this study, we present a novel four fold truncated double nested anti resonant hollow core fiber structure that addresses this challenge. Our 4T-DNANF enables greater control over phase-matching between core modes and air modes in the cladding, allowing for minimized FM loss and substantially increased HOM loss. Experimentally, we fabricated several HCFs: one with an FM loss of 0.1 dB/km and an HOM loss of 430 dB/km, and another with an FM loss of 0.13 dB/km with a HOM loss of 6500 dB/km, resulting in a higher-order mode extinction ratio of 50,000.

physics.optics↗

arXiv2024

Electrolyte spraying within H$_2$ bubbles during water electrolysis

Electrolytically generated gas bubbles can significantly hamper the overall electrolysis efficiency. Therefore it is crucial to understand their dynamics in order to optimise water electrolyzer systems. Here we demonstrate a distinct transport mechanism where coalescence with microbubbles drives electrolyte droplets, resulting from the fragmentation of the Worthington jet, into the gas phase during hydrogen evolution reaction, both in normal and microgravity environments. This indicates that the H$_2$ bubble is not only composed of hydrogen gas and vapor but also includes electrolyte fractions. Reminiscent of bursting bubbles on a liquid-gas interface, this behavior results in a flow inside the bubble, which is further affected by Marangoni convection at the gas-electrolyte interface, highlighting interface mobility. In the case of electrode-attached bubbles, the sprayed droplets form electrolyte puddles at the bubble-electrode contact area, affecting the dynamics near the three-phase contact line and favoring bubble detachment from the electrode. The results of this work unravel important insights into the physicochemical aspects of electrolytic gas bubbles, integral for optimizing gas-evolving electrochemical systems. Besides, our findings are essential for studying the limits of jet formation and rupture relevant to acid mist formation in electrowinning, generation of sea spray aerosols, impact of droplets on liquid surfaces, etc.

physics.flu-dyn↗

arXiv2024

Navigation-grade interferometric air-core antiresonant fibre optic gyroscope with enhanced thermal stability

We present a groundbreaking navigation-grade interferometric air-core fibre optic gyroscope (IFOG) using a quadrupolar-wound coil of four-tube truncated double nested antiresonant nodeless fibre (tDNANF). This state-of-the-art tDNANF simultaneously achieves low loss, low bend loss, single-spatial-mode operation, and exceptional linear polarization purity over a broad wavelength range. Our 469 m tDNANF coil demonstrated a polarization extinction ratio (PER) of ~20 dB when illuminated by an amplified spontaneous emission (ASE) source spanning 1525-1565 nm. Under these conditions, the gyro archives an angular random walk (ARW) of 0.0038 deg h-1/2 and a bias-stability (BS) drift over 8500 s of 0.0014 deg h-1, marking the first instance of navigation-grade performance in air-core FOGs. Additionally, we validated the low thermal sensitivity of air-core FOGs, with reductions of 9.24/10.68/6.82 compared to that of conventional polarization-maintaining solid-core FOGs of the same size across various temperature ranges. These results represent a significant step towards long-standing promise of high-precision inertial navigation applications with superior environmental adaptability.

physics.optics↗

arXiv2024

Origin and Customization of Bandgap in Chiral Phononic Crystals

The wave equation governing the wave propagation in chiral phononic crystals, established through force equilibrium law, conceals the underlying physical information. This has led to a controversy over the bandgap mechanism. In this letter, we theoretically unveil the reason of this controversy, and put forward an alternative approach from wave behavior to formulate the wave equation, offering a new pathway to articulate the bandgap physics directly. We identify the obstacles in coupled acoustic and optic branches to widen and lower the bandgap, and introduce an approach based on spherical hinges to decrease the barriers, for customizing the bandgap frequency and width. Finally, we validate our proposal through numerical simulation and experimental demonstration.

physics.app-ph↗

arXiv2023

Alignment is not sufficient to prevent large language models from generating harmful information: A psychoanalytic perspective

Large Language Models (LLMs) are central to a multitude of applications but struggle with significant risks, notably in generating harmful content and biases. Drawing an analogy to the human psyche's conflict between evolutionary survival instincts and societal norm adherence elucidated in Freud's psychoanalysis theory, we argue that LLMs suffer a similar fundamental conflict, arising between their inherent desire for syntactic and semantic continuity, established during the pre-training phase, and the post-training alignment with human values. This conflict renders LLMs vulnerable to adversarial attacks, wherein intensifying the models' desire for continuity can circumvent alignment efforts, resulting in the generation of harmful information. Through a series of experiments, we first validated the existence of the desire for continuity in LLMs, and further devised a straightforward yet powerful technique, such as incomplete sentences, negative priming, and cognitive dissonance scenarios, to demonstrate that even advanced LLMs struggle to prevent the generation of harmful information. In summary, our study uncovers the root of LLMs' vulnerabilities to adversarial attacks, hereby questioning the efficacy of solely relying on sophisticated alignment methods, and further advocates for a new training idea that integrates modal concepts alongside traditional amodal concepts, aiming to endow LLMs with a more nuanced understanding of real-world contexts and ethical considerations.

cs.CL↗

arXiv2023

Dual Adversarial Alignment for Realistic Support-Query Shift Few-shot Learning

Support-query shift few-shot learning aims to classify unseen examples (query set) to labeled data (support set) based on the learned embedding in a low-dimensional space under a distribution shift between the support set and the query set. However, in real-world scenarios the shifts are usually unknown and varied, making it difficult to estimate in advance. Therefore, in this paper, we propose a novel but more difficult challenge, RSQS, focusing on Realistic Support-Query Shift few-shot learning. The key feature of RSQS is that the individual samples in a meta-task are subjected to multiple distribution shifts in each meta-task. In addition, we propose a unified adversarial feature alignment method called DUal adversarial ALignment framework (DuaL) to relieve RSQS from two aspects, i.e., inter-domain bias and intra-domain variance. On the one hand, for the inter-domain bias, we corrupt the original data in advance and use the synthesized perturbed inputs to train the repairer network by minimizing distance in the feature level. On the other hand, for intra-domain variance, we proposed a generator network to synthesize hard, i.e., less similar, examples from the support set in a self-supervised manner and introduce regularized optimal transportation to derive a smooth optimal transportation plan. Lastly, a benchmark of RSQS is built with several state-of-the-art baselines among three datasets (CIFAR100, mini-ImageNet, and Tiered-Imagenet). Experiment results show that DuaL significantly outperforms the state-of-the-art methods in our benchmark.

cs.CV↗

arXiv2023

Self-Supervised Object Goal Navigation with In-Situ Finetuning

A household robot should be able to navigate to target objects without requiring users to first annotate everything in their home. Most current approaches to object navigation do not test on real robots and rely solely on reconstructed scans of houses and their expensively labeled semantic 3D meshes. In this work, our goal is to build an agent that builds self-supervised models of the world via exploration, the same as a child might - thus we (1) eschew the expense of labeled 3D mesh and (2) enable self-supervised in-situ finetuning in the real world. We identify a strong source of self-supervision (Location Consistency - LocCon) that can train all components of an ObjectNav agent, using unannotated simulated houses. Our key insight is that embodied agents can leverage location consistency as a self-supervision signal - collecting images from different views/angles and applying contrastive learning. We show that our agent can perform competitively in the real world and simulation. Our results also indicate that supervised training with 3D mesh annotations causes models to learn simulation artifacts, which are not transferrable to the real world. In contrast, our LocCon shows the most robust transfer in the real world among the set of models we compare to, and that the real-world performance of all models can be further improved with self-supervised LocCon in-situ training.

cs.RO↗

arXiv2023

Generalized Category Discovery with Decoupled Prototypical Network

Generalized Category Discovery (GCD) aims to recognize both known and novel categories from a set of unlabeled data, based on another dataset labeled with only known categories. Without considering differences between known and novel categories, current methods learn about them in a coupled manner, which can hurt model's generalization and discriminative ability. Furthermore, the coupled training approach prevents these models transferring category-specific knowledge explicitly from labeled data to unlabeled data, which can lose high-level semantic information and impair model performance. To mitigate above limitations, we present a novel model called Decoupled Prototypical Network (DPN). By formulating a bipartite matching problem for category prototypes, DPN can not only decouple known and novel categories to achieve different training targets effectively, but also align known categories in labeled and unlabeled data to transfer category-specific knowledge explicitly and capture high-level semantics. Furthermore, DPN can learn more discriminative features for both known and novel categories through our proposed Semantic-aware Prototypical Learning (SPL). Besides capturing meaningful semantic information, SPL can also alleviate the noise of hard pseudo labels through semantic-weighted soft assignment. Extensive experiments show that DPN outperforms state-of-the-art models by a large margin on all evaluation metrics across multiple benchmark datasets. Code and data are available at https://github.com/Lackel/DPN.

cs.CL↗

arXiv2023

Significant Light Extraction and Power Efficiency Enhancement of Organic Light Emitting Diodes by Subwavelength Dielectric-Nanomesh Using Large-area Nanoimprint

To improve the power efficiency of light emitting diodes (OLEDs), we developed a novel OLED structure, termed Dielectric-Nanomesh OLED (DNM-OLED), fabricated by large-area nanoimprint lithography (NIL). A dielectric-nanomesh substrate with a subwavelength nanomesh grating patterned into glass releases the photons trapped in ITO and organic layers by guided modes into leaky modes. And the dielectric-nanomesh substrate modifies the entire OLED layers into a corrugated configuration, which further lowers the parasitic resistance and hence the operating voltage. For a red phosphorescent OLED in our demonstration, with the DNM-OLED structure, the light extraction efficiency was enhanced by 1.33 fold and the operating voltage was lowered by 46%, acting together, leading to a 1.8-fold enhancement in power efficiency. Besides red OELDs, the DNM-OLED structure was also shown to be able to work for green and blue OELDs and achieved enhanced light extraction efficiency (1.31 fold for green light, 1.45 fold for blue light) without further structure modification, which demonstrated the light extraction enhancement by the DNM-OLED is broadband.

physics.optics↗

arXiv2023

Over 100% Light Extraction Enhancement of Organic Light Emitting Diodes Using Flat Moire Micro-Lens Array Fabricated by Double Nanoimprint Lithography Over a Large Area

To improve the light extraction efficiency of organic light emitting diodes (OLEDs), we developed a novel substrate, i.e., a metamaterial based flat Moire micro-lens array formed using double nanoimprint, termed Mlens-array, consisting of a hexagonal moiré pattern pillar array. By choosing a low refractive index dielectric material for the pillar array and a high refractive index dielectric material for the layer capped on top of the pillar array, we fabricated the Mlens-array behaving as a conventional convex optical micro-lens array. The Mlens-array was fabricated on a 4-inch wafer-size glass substrate by double-cycle compositional nanoimprint lithography (NIL) which is easy for achieving high throughput fabrication in large-scale. Applying the Mlens-array substrate in a typical green-emitting OLED, the light extraction efficiency was enhanced by over 100% (2.08-fold) compared to a control device fabricated on the conventional planar glass substrate.

physics.optics↗

arXiv2023

Enhancing Light Extraction of Organic Light Emitting Diodes by Deep-Groove High-index Dielectric Nanomesh Using Large-area Nanoimprint

To solve the conventional conflict between maintaining good charge transport property and achieving high light extraction efficiency when using micro/nanostructure patterned substrates to extract light from organic light emitting diodes (OLEDs), we developed a novel OLED structure, termed High-index Deep-Groove Dielectric Nanomesh OLED (HDNM-OLED), fabricated by large-area nanoimprint lithography (NIL). The key component is a nanostructure-patterned substrate embedded with a high-index deep-groove nanomesh and capped with a low-index planarization layer. The high-index and deep-groove nanomesh efficiently releases the tapped photons to achieve significantly enhanced light extraction. And the planarization layer helps to maintain the good charge transport property of the organic active layers deposited on top of it. For a green phosphorescent OLED in our demonstration, with the HDNM-OLED structure, compared to planar conventional ITO-OLED structure, the external quantum efficiency (EQE) was enhanced by 1.85-fold from 26% to 48% and power efficiency was enhanced by 1.86-fold from 42lm/W to 79lm/W. Besides green OELDs, the HDNM-OLED structure was also shown to be able to work for red and blue-emitting OELDs and achieved enhanced light extraction efficiency (1.58-fold for red light, 1.86-fold for blue light) without further structure modification, which demonstrated the light extraction enhancement by the HDNM-OLED is broadband.

physics.optics↗

arXiv2022

Multiscale multimesh finite element method | $\text{M}^2$-FEM: Hierarchical mesh-decoupling for integral structural theories

This study presents a generalized multiscale multimesh finite element method ($\text{M}^2$-FEM) that addresses several long-standing challenges in the numerical simulation of integral structural theories, often used to model multiscale and nonlocal effects. The major challenges in the numerical simulation of integral boundary value problems are primarily rooted in the coupling of the spatial discretization of the global (parent) and integral (child) domains which severely restricts the computational efficiency of existing algorithms by imposing an implicit trade-off in the accuracy achieved by the child domain and in the resources dedicated to the simulation of the overall parent domain. One of the most defining contributions of this study consists in the development of a mesh-decoupling technique that generates isolated sets of meshes such that the parent and child domains can be discretized and approximated independently. This mesh-decoupling has a multi-fold impact on the simulation of integral theories such that, when compared to existing state-of-the-art techniques, the proposed algorithm achieves simultaneously better numerical accuracy and efficiency (hence allowing a greater flexibility in both mesh size and computational cost trade-off decisions), greater ability to adopt generalized integral kernel functions, and the ability to handle non-regular (non-rectangular) domains via unstructured meshing. In this study, we choose a benchmark problem based on an extended version of the Eringen's nonlocal elasticity theory (implicitly, a multiscale theory) that leverages the use of generalized attenuation kernels and non-constant horizons of nonlocality. Nonetheless, the proposed $\text{M}^2$-FEM algorithm is very general and it can be applied to a variety of integral theories, even beyond structural elasticity.

math.NA↗