arXiv subjects

Di Wang

Publications and source records attributed to Di Wang.

At least 397 records · Page 22Linked to original sources

arXiv2024

Fair Text-to-Image Diffusion via Fair Mapping

In this paper, we address the limitations of existing text-to-image diffusion models in generating demographically fair results when given human-related descriptions. These models often struggle to disentangle the target language context from sociocultural biases, resulting in biased image generation. To overcome this challenge, we propose Fair Mapping, a flexible, model-agnostic, and lightweight approach that modifies a pre-trained text-to-image diffusion model by controlling the prompt to achieve fair image generation. One key advantage of our approach is its high efficiency. It only requires updating an additional linear network with few parameters at a low computational cost. By developing a linear network that maps conditioning embeddings into a debiased space, we enable the generation of relatively balanced demographic results based on the specified text condition. With comprehensive experiments on face image generation, we show that our method significantly improves image generation fairness with almost the same image quality compared to conventional diffusion models when prompted with descriptions related to humans. By effectively addressing the issue of implicit language bias, our method produces more fair and diverse image outputs.

cs.CV↗

arXiv2024

MoRAL: MoE Augmented LoRA for LLMs' Lifelong Learning

Adapting large language models (LLMs) to new domains/tasks and enabling them to be efficient lifelong learners is a pivotal challenge. In this paper, we propose MoRAL, i.e., Mixture-of-Experts augmented Low-Rank Adaptation for Lifelong Learning. MoRAL combines the multi-tasking abilities of MoE with the fine-tuning abilities of LoRA for effective life-long learning of LLMs. In contrast to the conventional approaches that use factual triplets as inputs MoRAL relies on simple question-answer pairs, which is a more practical and effective strategy for robust and efficient learning. Owing to new data settings, we introduce a new evaluation benchmark namely: Life Long Learning of LLM (5L-bench) encompassing a newly curated dataset of question-answer pairs, and a set of evaluation metrics for rigorous evaluation of MoRAL in open-book and closed-book settings. Experimental evaluation shows (i) LLMs learn fast in open-book settings with up to 30.15% improvement in "RA" for Phi-2-2.7B compared to closed-book (for models fine-tuned with MoRAL); (ii) MoRAL shows higher performance improvement for models with a greater number of parameters; (iii) MoRAL is robust to catastrophic forgetting offering better knowledge retention compared to baselines.

cs.CL↗

arXiv2024

Near-perfect Coverage Manifold Estimation in Cellular Networks via conditional GAN

This paper presents a conditional generative adversarial network (cGAN) that translates base station location (BSL) information of any Region-of-Interest (RoI) to location-dependent coverage probability values within a subset of that region, called the region-of-evaluation (RoE). We train our network utilizing the BSL data of India, the USA, Germany, and Brazil. In comparison to the state-of-the-art convolutional neural networks (CNNs), our model improves the prediction error ($L_1$ difference between the coverage manifold generated by the network under consideration and that generated via simulation) by two orders of magnitude. Moreover, the cGAN-generated coverage manifolds appear to be almost visually indistinguishable from the ground truth.

cs.NI↗

arXiv2024

Theoretical Analysis of Robust Overfitting for Wide DNNs: An NTK Approach

Adversarial training (AT) is a canonical method for enhancing the robustness of deep neural networks (DNNs). However, recent studies empirically demonstrated that it suffers from robust overfitting, i.e., a long time AT can be detrimental to the robustness of DNNs. This paper presents a theoretical explanation of robust overfitting for DNNs. Specifically, we non-trivially extend the neural tangent kernel (NTK) theory to AT and prove that an adversarially trained wide DNN can be well approximated by a linearized DNN. Moreover, for squared loss, closed-form AT dynamics for the linearized DNN can be derived, which reveals a new AT degeneration phenomenon: a long-term AT will result in a wide DNN degenerates to that obtained without AT and thus cause robust overfitting. Based on our theoretical results, we further design a method namely Adv-NTK, the first AT algorithm for infinite-width DNNs. Experiments on real-world datasets show that Adv-NTK can help infinite-width DNNs enhance comparable robustness to that of their finite-width counterparts, which in turn justifies our theoretical findings. The code is available at https://github.com/fshp971/adv-ntk.

cs.LG↗

arXiv2024

Tidal disruption event associated with the quasi-periodic eruptions from GSN 069: Possible disruption of a common envelope

X-ray quasi-periodic eruptions (QPEs) from the galactic nucleus have been found in several galaxies. Among them, GSN 069 is the only one with a tidal disruption event (TDE), which was recently found to have brightened again 9 years after the main outburst. However, the origin of this TDE is still unclear. This Letter explores a particular model for the TDE. By comparing the fallback time with observations, we found the TDE could not be caused by the disruption of the envelope of a single star in the tidal stripping model. Thus, we suggest that it is a disruption of a common envelope (CE). By calculating the fallback rate of such a model, we reproduced the second peak in the observed TDE light curve. If this model is correct, this TDE will be the closest observation to a direct observation of CE, which has never been directly observed.

astro-ph.HE↗

arXiv2024

Decomposition-Based Synthesis for Applying Divide-and-Conquer-Like Algorithmic Paradigms

Algorithmic paradigms such as divide-and-conquer (D&C) are proposed to guide developers in designing efficient algorithms, but it can still be difficult to apply algorithmic paradigms to practical tasks. To ease the usage of paradigms, many research efforts have been devoted to the automatic application of algorithmic paradigms. However, most existing approaches to this problem rely on syntax-based program transformations and thus put significant restrictions on the original program. In this paper, we study the automatic application of D&C and several similar paradigms, denoted as D&C-like algorithmic paradigms, and aim to remove the restrictions from syntax-based transformations. To achieve this goal, we propose an efficient synthesizer, named AutoLifter, which does not depend on syntax-based transformations. Specifically, the main challenge of applying algorithmic paradigms is from the large scale of the synthesized programs, and AutoLifter addresses this challenge by applying two novel decomposition methods that do not depend on the syntax of the input program, component elimination and variable elimination, to soundly divide the whole problem into simpler subtasks, each synthesizing a sub-program of the final program and being tractable with existing synthesizers. We evaluate AutoLifter on 96 programming tasks related to 6 different algorithmic paradigms. AutoLifter solves 82/96 tasks with an average time cost of 20.17 seconds, significantly outperforming existing approaches.

cs.PL↗

arXiv2024

Unitary and efficient spin squeezing in cavity optomechanics

We propose an approach to produce spin squeezed states of a large number of nitrogen-vacancy centers in diamond nanostructures coupled to an optical cavity. Unlike the previous squeezing method proposed by Bennett et al. [Phys. Rev. Lett. 110, 156402 (2013)], which is limited by phonon number fluctuations due to the existence of phonon-spin entanglement, our proposal can completely erase the entanglement between spins and hybrid phonon-photon mode mediating the effective spin-spin interaction, and thus achieves unitary one-axis-twisting interactions between nitrogen-vacancy centres, yielding a squeezing scaling $J^{-2/3}$, where J is the total angular momentum. We found that, under certain conditions, our method has the potential to enhance the spin-spin nonlinear interactions. We also proposed a scheme utilizing repeatedly applying the one-axis-twisting evolution to two orthogonal spin directions, which enables the transformation of the one-axis-twisting interactions into two-axis-twisting type, and therefore leads to the spin squeezing with Heisenberg-limited scaling $J^{-1}$. Taking into account the noise effects of spin dephasing and relaxtion, we found that the proposed approaches are robust against imperfections.

quant-ph↗

arXiv2024

Conformal Loss-Controlling Prediction

Conformal prediction is a learning framework controlling prediction coverage of prediction sets, which can be built on any learning algorithm for point prediction. This work proposes a learning framework named conformal loss-controlling prediction, which extends conformal prediction to the situation where the value of a loss function needs to be controlled. Different from existing works about risk-controlling prediction sets and conformal risk control with the purpose of controlling the expected values of loss functions, the proposed approach in this paper focuses on the loss for any test object, which is an extension of conformal prediction from miscoverage loss to some general loss. The controlling guarantee is proved under the assumption of exchangeability of data in finite-sample cases and the framework is tested empirically for classification with a class-varying loss and statistical postprocessing of numerical weather forecasting applications, which are introduced as point-wise classification and point-wise regression problems. All theoretical analysis and experimental results confirm the effectiveness of our loss-controlling approach.

cs.LG↗

arXiv2024

Loss-Controlling Calibration for Predictive Models

We propose a learning framework for calibrating predictive models to make loss-controlling prediction for exchangeable data, which extends our recently proposed conformal loss-controlling prediction for more general cases. By comparison, the predictors built by the proposed loss-controlling approach are not limited to set predictors, and the loss function can be any measurable function without the monotone assumption. To control the loss values in an efficient way, we introduce transformations preserving exchangeability to prove finite-sample controlling guarantee when the test label is obtained, and then develop an approximation approach to construct predictors. The transformations can be built on any predefined function, which include using optimization algorithms for parameter searching. This approach is a natural extension of conformal loss-controlling prediction, since it can be reduced to the latter when the set predictors have the nesting property and the loss functions are monotone. Our proposed method is applied to selective regression and high-impact weather forecasting problems, which demonstrates its effectiveness for general loss-controlling prediction.

cs.LG↗

arXiv2024

NIV-SSD: Neighbor IoU-Voting Single-Stage Object Detector From Point Cloud

Previous single-stage detectors typically suffer the misalignment between localization accuracy and classification confidence. To solve the misalignment problem, we introduce a novel rectification method named neighbor IoU-voting (NIV) strategy. Typically, classification and regression are treated as separate branches, making it challenging to establish a connection between them. Consequently, the classification confidence cannot accurately reflect the regression quality. NIV strategy can serve as a bridge between classification and regression branches by calculating two types of statistical data from the regression output to correct the classification confidence. Furthermore, to alleviate the imbalance of detection accuracy for complete objects with dense points (easy objects) and incomplete objects with sparse points (difficult objects), we propose a new data augmentation scheme named object resampling. It undersamples easy objects and oversamples difficult objects by randomly transforming part of easy objects into difficult objects. Finally, combining the NIV strategy and object resampling augmentation, we design an efficient single-stage detector termed NIV-SSD. Extensive experiments on several datasets indicate the effectiveness of the NIV strategy and the competitive performance of the NIV-SSD detector. The code will be available at https://github.com/Say2L/NIV-SSD.

cs.CV↗

arXiv2024

Communication Efficient and Provable Federated Unlearning

We study federated unlearning, a novel problem to eliminate the impact of specific clients or data points on the global model learned via federated learning (FL). This problem is driven by the right to be forgotten and the privacy challenges in FL. We introduce a new framework for exact federated unlearning that meets two essential criteria: \textit{communication efficiency} and \textit{exact unlearning provability}. To our knowledge, this is the first work to tackle both aspects coherently. We start by giving a rigorous definition of \textit{exact} federated unlearning, which guarantees that the unlearned model is statistically indistinguishable from the one trained without the deleted data. We then pinpoint the key property that enables fast exact federated unlearning: total variation (TV) stability, which measures the sensitivity of the model parameters to slight changes in the dataset. Leveraging this insight, we develop a TV-stable FL algorithm called \texttt{FATS}, which modifies the classical \texttt{\underline{F}ed\underline{A}vg} algorithm for \underline{T}V \underline{S}tability and employs local SGD with periodic averaging to lower the communication round. We also design efficient unlearning algorithms for \texttt{FATS} under two settings: client-level and sample-level unlearning. We provide theoretical guarantees for our learning and unlearning algorithms, proving that they achieve exact federated unlearning with reasonable convergence rates for both the original and unlearned models. We empirically validate our framework on 6 benchmark datasets, and show its superiority over state-of-the-art methods in terms of accuracy, communication cost, computation cost, and unlearning efficacy.

cs.LG↗

arXiv2024

Distilling Autoregressive Models to Obtain High-Performance Non-Autoregressive Solvers for Vehicle Routing Problems with Faster Inference Speed

Neural construction models have shown promising performance for Vehicle Routing Problems (VRPs) by adopting either the Autoregressive (AR) or Non-Autoregressive (NAR) learning approach. While AR models produce high-quality solutions, they generally have a high inference latency due to their sequential generation nature. Conversely, NAR models generate solutions in parallel with a low inference latency but generally exhibit inferior performance. In this paper, we propose a generic Guided Non-Autoregressive Knowledge Distillation (GNARKD) method to obtain high-performance NAR models having a low inference latency. GNARKD removes the constraint of sequential generation in AR models while preserving the learned pivotal components in the network architecture to obtain the corresponding NAR models through knowledge distillation. We evaluate GNARKD by applying it to three widely adopted AR models to obtain NAR VRP solvers for both synthesized and real-world instances. The experimental results demonstrate that GNARKD significantly reduces the inference time (4-5 times faster) with acceptable performance drop (2-3\%). To the best of our knowledge, this study is first-of-its-kind to obtain NAR VRP solvers from AR ones through knowledge distillation.

cs.LG↗

arXiv2024

Antonym vs Synonym Distinction using InterlaCed Encoder NETworks (ICE-NET)

Antonyms vs synonyms distinction is a core challenge in lexico-semantic analysis and automated lexical resource construction. These pairs share a similar distributional context which makes it harder to distinguish them. Leading research in this regard attempts to capture the properties of the relation pairs, i.e., symmetry, transitivity, and trans-transitivity. However, the inability of existing research to appropriately model the relation-specific properties limits their end performance. In this paper, we propose InterlaCed Encoder NETworks (i.e., ICE-NET) for antonym vs synonym distinction, that aim to capture and model the relation-specific properties of the antonyms and synonyms pairs in order to perform the classification task in a performance-enhanced manner. Experimental evaluation using the benchmark datasets shows that ICE-NET outperforms the existing research by a relative score of upto 1.8% in F1-measure. We release the codes for ICE-NET at https://github.com/asif6827/ICENET.

cs.CL↗

arXiv2024

Weighted Spectral Filters for Kernel Interpolation on Spheres: Estimates of Prediction Accuracy for Noisy Data

Spherical radial-basis-based kernel interpolation abounds in image sciences including geophysical image reconstruction, climate trends description and image rendering due to its excellent spatial localization property and perfect approximation performance. However, in dealing with noisy data, kernel interpolation frequently behaves not so well due to the large condition number of the kernel matrix and instability of the interpolation process. In this paper, we introduce a weighted spectral filter approach to reduce the condition number of the kernel matrix and then stabilize kernel interpolation. The main building blocks of the proposed method are the well developed spherical positive quadrature rules and high-pass spectral filters. Using a recently developed integral operator approach for spherical data analysis, we theoretically demonstrate that the proposed weighted spectral filter approach succeeds in breaking through the bottleneck of kernel interpolation, especially in fitting noisy data. We provide optimal approximation rates of the new method to show that our approach does not compromise the predicting accuracy. Furthermore, we conduct both toy simulations and two real-world data experiments with synthetically added noise in geophysical image reconstruction and climate image processing to verify our theoretical assertions and show the feasibility of the weighted spectral filter approach.

cs.LG↗

arXiv2024

Truth Forest: Toward Multi-Scale Truthfulness in Large Language Models through Intervention without Tuning

Despite the great success of large language models (LLMs) in various tasks, they suffer from generating hallucinations. We introduce Truth Forest, a method that enhances truthfulness in LLMs by uncovering hidden truth representations using multi-dimensional orthogonal probes. Specifically, it creates multiple orthogonal bases for modeling truth by incorporating orthogonal constraints into the probes. Moreover, we introduce Random Peek, a systematic technique considering an extended range of positions within the sequence, reducing the gap between discerning and generating truth features in LLMs. By employing this approach, we improved the truthfulness of Llama-2-7B from 40.8\% to 74.5\% on TruthfulQA. Likewise, significant improvements are observed in fine-tuned models. We conducted a thorough analysis of truth features using probes. Our visualization results show that orthogonal probes capture complementary truth-related features, forming well-defined clusters that reveal the inherent structure of the dataset.

cs.CL↗

arXiv2024

Deterministic Near-Linear Time Minimum Cut in Weighted Graphs

In 1996, Karger [Kar96] gave a startling randomized algorithm that finds a minimum-cut in a (weighted) graph in time $O(m\log^3n)$ which he termed near-linear time meaning linear (in the size of the input) times a polylogarthmic factor. In this paper, we give the first deterministic algorithm which runs in near-linear time for weighted graphs. Previously, the breakthrough results of Kawarabayashi and Thorup [KT19] gave a near-linear time algorithm for simple graphs. The main technique here is a clustering procedure that perfectly preserves minimum cuts. Recently, Li [Li21] gave an $m^{1+o(1)}$ deterministic minimum-cut algorithm for weighted graphs; this form of running time has been termed "almost-linear''. Li uses almost-linear time deterministic expander decompositions which do not perfectly preserve minimum cuts, but he can use these clusterings to, in a sense, "derandomize'' the methods of Karger. In terms of techniques, we provide a structural theorem that says there exists a sparse clustering that preserves minimum cuts in a weighted graph with $o(1)$ error. In addition, we construct it deterministically in near linear time. This was done exactly for simple graphs in [KT19, HRW20] and with polylogarithmic error for weighted graphs in [Li21]. Extending the techniques in [KT19, HRW20] to weighted graphs presents significant challenges, and moreover, the algorithm can only polylogarithmically approximately preserve minimum cuts. A remaining challenge is to reduce the polylogarithmic-approximate clusterings to $1+o(1/\log n)$-approximate so that they can be applied recursively as in [Li21] over $O(\log n)$ many levels. This is an additional challenge that requires building on properties of tree-packings in the presence of a wide range of edge weights to, for example, find sources for local flow computations which identify minimum cuts that cross clusters.

cs.DS↗

arXiv2024

A comprehensive framework for occluded human pose estimation

Occlusion presents a significant challenge in human pose estimation. The challenges posed by occlusion can be attributed to the following factors: 1) Data: The collection and annotation of occluded human pose samples are relatively challenging. 2) Feature: Occlusion can cause feature confusion due to the high similarity between the target person and interfering individuals. 3) Inference: Robust inference becomes challenging due to the loss of complete body structural information. The existing methods designed for occluded human pose estimation usually focus on addressing only one of these factors. In this paper, we propose a comprehensive framework DAG (Data, Attention, Graph) to address the performance degradation caused by occlusion. Specifically, we introduce the mask joints with instance paste data augmentation technique to simulate occlusion scenarios. Additionally, an Adaptive Discriminative Attention Module (ADAM) is proposed to effectively enhance the features of target individuals. Furthermore, we present the Feature-Guided Multi-Hop GCN (FGMP-GCN) to fully explore the prior knowledge of body structure and improve pose estimation results. Through extensive experiments conducted on three benchmark datasets for occluded human pose estimation, we demonstrate that the proposed method outperforms existing methods. Code and data will be publicly available.

cs.CV↗

arXiv2023

Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks

Graph convolution networks (GCNs) are extensively utilized in various graph tasks to mine knowledge from spatial data. Our study marks the pioneering attempt to quantitatively investigate the GCN robustness over omnipresent heterophilic graphs for node classification. We uncover that the predominant vulnerability is caused by the structural out-of-distribution (OOD) issue. This finding motivates us to present a novel method that aims to harden GCNs by automatically learning Latent Homophilic Structures over heterophilic graphs. We term such a methodology as LHS. To elaborate, our initial step involves learning a latent structure by employing a novel self-expressive technique based on multi-node interactions. Subsequently, the structure is refined using a pairwisely constrained dual-view contrastive learning approach. We iteratively perform the above procedure, enabling a GCN model to aggregate information in a homophilic way on heterophilic graphs. Armed with such an adaptable structure, we can properly mitigate the structural OOD threats over heterophilic graphs. Experiments on various benchmarks show the effectiveness of the proposed LHS approach for robust GCNs.

cs.LG↗