Distilling Large Language Models for Text-Attributed Graph Learning

Text-Attributed Graphs (TAGs) are graphs of connected textual documents. Graph models can efficiently learn TAGs, but their training heavily relies on human-annotated labels, which are scarce or even unavailable in many applications. Large language models (LLMs) have recently demonstrated remarkable capabilities in few-shot and zero-shot TAG learning, but they suffer from scalability, cost, and privacy issues. Therefore, in this work, we focus on synergizing LLMs and graph models with their complementary strengths by distilling the power of LLMs to a local graph model on TAG learning. To address the inherent gaps between LLMs (generative models for texts) and graph models (discriminative models for graphs), we propose first to let LLMs teach an interpreter with rich textual rationale and then let a student model mimic the interpreter's reasoning without LLMs' textual rationale. Extensive experiments validate the efficacy of our proposed framework.

LoRETTA: Low-Rank Economic Tensor-Train Adaptation for Ultra-Low-Parameter Fine-Tuning of Large Language Models

Various parameter-efficient fine-tuning (PEFT) techniques have been proposed to enable computationally efficient fine-tuning while maintaining model performance. However, existing PEFT methods are still limited by the growing number of trainable parameters with the rapid deployment of Large Language Models (LLMs). To address this challenge, we present LoRETTA, an ultra-parameter-efficient framework that significantly reduces trainable parameters through tensor-train decomposition. Specifically, we propose two methods, named {LoRETTA}$_{adp}$ and {LoRETTA}$_{rep}$. The former employs tensorized adapters, offering a high-performance yet lightweight approach for the fine-tuning of LLMs. The latter emphasizes fine-tuning via weight parameterization with a set of small tensor factors. LoRETTA achieves comparable or better performance than most widely used PEFT methods with up to $100\times$ fewer parameters on the LLaMA-2-7B models. Furthermore, empirical results demonstrate that the proposed method effectively improves training efficiency, enjoys better multi-task learning performance, and enhances the anti-overfitting capability. Plug-and-play codes built upon the Huggingface framework and PEFT library will be released.

Multi-Agent Generative Adversarial Interactive Self-Imitation Learning for AUV Formation Control and Obstacle Avoidance

Multiple autonomous underwater vehicles (multi-AUV) can cooperatively accomplish tasks that a single AUV cannot complete. Recently, multi-agent reinforcement learning has been introduced to control of multi-AUV. However, designing efficient reward functions for various tasks of multi-AUV control is difficult or even impractical. Multi-agent generative adversarial imitation learning (MAGAIL) allows multi-AUV to learn from expert demonstration instead of pre-defined reward functions, but suffers from the deficiency of requiring optimal demonstrations and not surpassing provided expert demonstrations. This paper builds upon the MAGAIL algorithm by proposing multi-agent generative adversarial interactive self-imitation learning (MAGAISIL), which can facilitate AUVs to learn policies by gradually replacing the provided sub-optimal demonstrations with self-generated good trajectories selected by a human trainer. Our experimental results in a multi-AUV formation control and obstacle avoidance task on the Gazebo platform with AUV simulator of our lab show that AUVs trained via MAGAISIL can surpass the provided sub-optimal expert demonstrations and reach a performance close to or even better than MAGAIL with optimal demonstrations. Further results indicate that AUVs' policies trained via MAGAISIL can adapt to complex and different tasks as well as MAGAIL learning from optimal demonstrations.

See the Unseen: Better Context-Consistent Knowledge-Editing by Noises

Knowledge-editing updates knowledge of large language models (LLMs) and contributes to the interpretability and application of LLMs. However, knowledge applying is context-consistent: LLMs can recall the same knowledge in different contexts. Existing works ignore this property and the editing lacks generalization. In this paper, we empirically find that the effects of different contexts upon LLMs in recalling the same knowledge follow a Gaussian-like distribution. We then sample Gaussian noises to simulate the effects of different contexts when updating LLMs. By such, we can make LLMs see the unseen contexts where the edited knowledge will be applied, therefore improving the editing generalization. Experimental results on three LLMs demonstrate the effectiveness of our methods and also distinguish our methods from the others of fine-tuning LLMs by noises.

Real-Time FJ/MAC PDE Solvers via Tensorized, Back-Propagation-Free Optical PINN Training

Solving partial differential equations (PDEs) numerically often requires huge computing time, energy cost, and hardware resources in practical applications. This has limited their applications in many scenarios (e.g., autonomous systems, supersonic flows) that have a limited energy budget and require near real-time response. Leveraging optical computing, this paper develops an on-chip training framework for physics-informed neural networks (PINNs), aiming to solve high-dimensional PDEs with fJ/MAC photonic power consumption and ultra-low latency. Despite the ultra-high speed of optical neural networks, training a PINN on an optical chip is hard due to (1) the large size of photonic devices, and (2) the lack of scalable optical memory devices to store the intermediate results of back-propagation (BP). To enable realistic optical PINN training, this paper presents a scalable method to avoid the BP process. We also employ a tensor-compressed approach to improve the convergence and scalability of our optical PINN training. This training framework is designed with tensorized optical neural networks (TONN) for scalable inference acceleration and MZI phase-domain tuning for \textit{in-situ} optimization. Our simulation results of a 20-dim HJB PDE show that our photonic accelerator can reduce the number of MZIs by a factor of $1.17\times 10^3$, with only $1.36$ J and $1.15$ s to solve this equation. This is the first real-size optical PINN training framework that can be applied to solve high-dimensional PDEs.

Dual-Functional Artificial Noise (DFAN) Aided Robust Covert Communications in Integrated Sensing and Communications

This paper investigates covert communications in an integrated sensing and communications system, where a dual-functional base station (called Alice) covertly transmits signals to a covert user (called Bob) while sensing multiple targets, with one of them acting as a potential watcher (called Willie) and maliciously eavesdropping on legitimate communications. To shelter the covert communications, Alice transmits additional dual-functional artificial noise (DFAN) with a varying power not only to create uncertainty at Willie's signal reception to confuse Willie but also to sense the targets simultaneously. Based on this framework, the weighted sum of the sensing beampattern means square error (MSE) and cross correlation is minimized by jointly optimizing the covert communication and DFAN signals subject to the minimum covert rate requirement. The robust design considers both cases of imperfect Willie's CSI (WCSI) and statistical WCSI. Under the worst-case assumption that Willie can adaptively adjust the detection threshold to achieve the best detection performance, the minimum detection error probability (DEP) at Willie is analytically derived in the closed-form expression. The formulated covertness constrained optimization problems are tackled by a feasibility-checking based difference-of-convex relaxation (DC) algorithm utilizing the S-procedure, Bernstein-type inequality, and the DC method. Simulation results validate the feasibility of the proposed scheme and demonstrate the covertness performance gains achieved by our proposed design over various benchmarks.

Dynamic MIMO Architecture Design for Near-Field Communications

A novel dynamic hybrid beamforming architecture is proposed to achieve the spatial multiplexing-power consumption tradeoff for near-field multiple-input multiple-output (MIMO) networks, where each radio frequency (RF) chain is connected to each antenna using a couple of independent phase shifters to reduce the number of required RF chains. Based on this architecture, an optimization problem is formulated that maximizes the sum of achievable rates while minimizing the hardware power consumption. Both continuous and discrete phase shifters are considered. 1) For continuous phase shifters, a weighted minimum mean-square error-based two-stage (WMMSE-TS) algorithm is proposed, where the same performance as the optimal fully-digital beamformer can be achieved by the proposed hybrid beamformer even if the number of RF chains equals the number of data streams. 2) For discrete phase shifters, a penalty-based layered iterative (PLI) algorithm is proposed. The closed-form analog and baseband digital beamformers are derived in each iteration. Simulation results demonstrate that: 1) the proposed dynamic beamforming architecture outperforms the conventional fixed hybrid beamforming architecture in terms of spatial multiplexing-power consumption tradeoff, and 2) the proposed algorithms achieve better performance than the other baseline schemes.

Zero-1-to-3: Domain-level Zero-shot Cognitive Diagnosis via One Batch of Early-bird Students towards Three Diagnostic Objectives

Cognitive diagnosis seeks to estimate the cognitive states of students by exploring their logged practice quiz data. It plays a pivotal role in personalized learning guidance within intelligent education systems. In this paper, we focus on an important, practical, yet often underexplored task: domain-level zero-shot cognitive diagnosis (DZCD), which arises due to the absence of student practice logs in newly launched domains. Recent cross-domain diagnostic models have been demonstrated to be a promising strategy for DZCD. These methods primarily focus on how to transfer student states across domains. However, they might inadvertently incorporate non-transferable information into student representations, thereby limiting the efficacy of knowledge transfer. To tackle this, we propose Zero-1-to-3, a domain-level zero-shot cognitive diagnosis framework via one batch of early-bird students towards three diagnostic objectives. Our approach initiates with pre-training a diagnosis model with dual regularizers, which decouples student states into domain-shared and domain-specific parts. The shared cognitive signals can be transferred to the target domain, enriching the cognitive priors for the new domain, which ensures the cognitive state propagation objective. Subsequently, we devise a strategy to generate simulated practice logs for cold-start students through analyzing the behavioral patterns from early-bird students, fulfilling the domain-adaption goal. Consequently, we refine the cognitive states of cold-start students as diagnostic outcomes via virtual data, aligning with the diagnosis-oriented goal. Finally, extensive experiments on six real-world datasets highlight the efficacy of our model for DZCD and its practical application in question recommendation.

Generate E-commerce Product Background by Integrating Category Commonality and Personalized Style

The state-of-the-art methods for e-commerce product background generation suffer from the inefficiency of designing product-wise prompts when scaling up the production, as well as the ineffectiveness of describing fine-grained styles when customizing personalized backgrounds for some specific brands. To address these obstacles, we integrate the category commonality and personalized style into diffusion models. Concretely, we propose a Category-Wise Generator to enable large-scale background generation for the first time. A unique identifier in the prompt is assigned to each category, whose attention is located on the background by a mask-guided cross attention layer to learn the category-wise style. Furthermore, for products with specific and fine-grained requirements in layout, elements, etc, a Personality-Wise Generator is devised to learn such personalized style directly from a reference image to resolve textual ambiguities, and is trained in a self-supervised manner for more efficient training data usage. To advance research in this field, the first large-scale e-commerce product background generation dataset BG60k is constructed, which covers more than 60k product images from over 2k categories. Experiments demonstrate that our method could generate high-quality backgrounds for different categories, and maintain the personalized background style of reference images. The link to BG60k and codes will be available soon.

Non-Euclidean Spatial Graph Neural Network

Spatial networks are networks whose graph topology is constrained by their embedded spatial space. Understanding the coupled spatial-graph properties is crucial for extracting powerful representations from spatial networks. Therefore, merely combining individual spatial and network representations cannot reveal the underlying interaction mechanism of spatial networks. Besides, existing spatial network representation learning methods can only consider networks embedded in Euclidean space, and can not well exploit the rich geometric information carried by irregular and non-uniform non-Euclidean space. In order to address this issue, in this paper we propose a novel generic framework to learn the representation of spatial networks that are embedded in non-Euclidean manifold space. Specifically, a novel message-passing-based neural network is proposed to combine graph topology and spatial geometry, where spatial geometry is extracted as messages on the edges. We theoretically guarantee that the learned representations are provably invariant to important symmetries such as rotation or translation, and simultaneously maintain sufficient ability in distinguishing different geometric structures. The strength of our proposed method is demonstrated through extensive experiments on both synthetic and real-world datasets.