40 accepted papers
Diffusion models often exhibit inconsistent sample quality due to stochastic variations inherent in their sampling trajectories. Although training-based fine-tuning and inference-time alignment techniques aim to improve sample fidelity, they typically necessitate full denoising processes and externa
Reward models (RMs) are crucial in reinforcement learning from human feedback (RLHF) to align large language models (LLMs) with human values. However, RM training data is commonly recognized as low-quality, always containing preference conflicts and inductive biases, such as response length or speak…
The efficiency of Bayesian optimization (BO) relies heavily on the choice of the Gaussian process (GP) kernel, which plays a central role in balancing exploration and exploitation under limited evaluation budgets. Traditional BO methods often rely on fixed or heuristic kernel selection strategies, w…
In this work, we focus on the alignment problem of diffusion models with a continuous reward function, which represents specific objectives for downstream tasks, such as increasing darkness or improving the aesthetics of images. The central goal of the alignment problem is to adjust the distribution…
This study investigates a networked integrated sensing and communication (ISAC) system, where a central processor coordinates multiple ISAC transmitters and a sensing receiver through capacity-limited fronthaul links. The system aims to serve multiple mobile users and simultaneously sense a point ta…
A common characteristic in integer linear programs (ILPs) is symmetry, allowing variables to be permuted without altering the underlying problem structure. Recently, GNNs have emerged as a promising approach for solving ILPs. However, a significant challenge arises when applying GNNs to ILPs with s…
Cooperative sensing, a technique employed in cognitive radio (CR) networks for spectrum sensing, exhibits promising potential in bolstering spectrum utilization and enhancing network performance. This approach leverages the information captured by distributed CR users, which is subsequently aggregat…
Diffusion models have emerged as state-of-the-art generative models for image generation. However, sampling from diffusion models is usually time-consuming due to the inherent autoregressive nature of their sampling process. In this work, we propose a novel approach that accelerates the sampling of…
In Federated Learning (FL), a framework to train machine learning models across distributed data, well-known algorithms like FedAvg tend to have slow convergence rates, resulting in high communication costs during training. To address this challenge, we introduce FedLion, an adaptive federated optim…
Acquiring accurate channel state information (CSI) is challenging in dynamic millimeter wave networks due to the excessive signaling overhead. In this work, we leverage the integrated sensing and communication (ISAC) technique for simultaneous data communication and CSI acquisition via proactive sen…
While distributed multi-cell resource allocation (D-MCRA) is promising for improving the spectral efficiency of cellular systems, it is challenging to realize in practice due to the large overhead for exchanging the channel state information (CSI) between base stations (BSs) and limited backhaul ban…
Integer linear programs (ILPs) are commonly employed to model diverse practical problems such as scheduling and planning. Recently, machine learning techniques have been utilized to solve ILPs. A straightforward idea is to train a model via supervised learning, with an ILP as the input and an opti…
In this study, we delve into an emerging optimization challenge involving a black-box objective function that can only be gauged via a ranking oracle—a situation frequently encountered in real-world scenarios, especially when the function is evaluated by human judges. A prominent instance of such a…
Federated Learning (FL) is a promising privacy-preserving distributed learning paradigm but suffers from high communi- cation cost when training large-scale machine learning models. Sign-based methods, such as SignSGD, have been proposed as a biased gradient compression technique for reducing the co…
Cross-domain named entity recognition (NER), aiming to address the limitation of labeled resources in the target domain, is a challenging yet important task. Most existing studies alleviate the data discrepancy across different domains at the coarse level via combing NER with language modelings or i…
Batch normalization (BN) has been widely used for accelerating the training of deep neural networks. However, recent findings show that, in the federated learning (FL) scenarios, BN can damage the learning performance when the clients have non-i.i.d. data. While several FL schemes have been proposed…
As a novel distributed learning paradigm, federated learning (FL) faces serious challenges in dealing with massive clients with heterogeneous data distribution and computation and communication resources. Various client-variance-reduction schemes and client sampling strategies have been respectively…
Active learning is a critical technique for reducing labelling load by selecting the most informative data. Most previous works applied active learning on Named Entity Recognition (token-level task) similar to the text classification (sentence-level task). They failed to consider the heterogeneity o…
Grammatical error correction (GEC) is a promising task aimed at correcting errors in a text. Many methods have been proposed to facilitate this task with remarkable results. However, most of them only focus on enhancing textual feature extraction without exploring the usage of other modalities’ info…
The impression is crucial for the referring physicians to grasp key information since it is concluded from the findings and reasoning of radiologists. To alleviate the workload of radiologists and reduce repetitive human labor in impression writing, many researchers have focused on automatic impress…
In this paper, we consider the joint beamforming design for simultaneous sensing and communication in a wireless multi-user system. Different from the existing works that mostly are for single target, we consider sensing the channel parameters of multiple targets while communicating with multiple us…
We study a matrix recovery problem with unknown correspondence: given the observation matrix $M_o=[A,\tilde P B]$, where $\tilde P$ is an unknown permutation matrix, we aim to recover the underlying matrix $M=[A,B]$. Such problem commonly arises in many applications where heterogeneous data are util…
To cope with the bottlenecks of the high computational complexity and excessive inter-connection communication in the conventional centralized baseband processing architecture, the decentralized baseband processing (DBP) architecture has been proposed, where the antennas are partitioned into multipl…
Cross-domain named entity recognition (NER) aims to borrow the entity information from the source domain to help the entity recognition in the target domain with limited labeled data. Despite the promising performance of existing approaches, most of them focus on reducing the discrepancy of token re…
Few-Shot Relation Extraction aims at predicting the relation for a pair of entities in a sentence by training with a few labelled examples in each relation. Some recent works have introduced relation information (i.e., relation labels or descriptions) to assist model learning based on Prototype Netw…
The impression section of a radiology report summarizes the most prominent observation from the findings section and is the most important section for radiologists to communicate to physicians. Summarizing findings is time-consuming and can be prone to error for inexperienced radiologists, and thus…
Named entity recognition (NER) is a fundamental and important task in NLP, aiming at identifying named entities (NEs) from free text. Recently, since the multi-head attention mechanism applied in the Transformer model can effectively capture longer contextual information, Transformer-based models ha…
Few-shot Relation Extraction refers to fast adaptation to novel relation classes with few samples through training on the known relation classes. Most existing methods focus on implicitly introducing relation information (i.e., relation label or relation description) to constrain the prototype repre…
Federated learning (FL) is encountered with the challenge of training a model in massive and heterogeneous networks. Model averaging (MA) has become a popular FL paradigm where parallel (stochastic) gradient descent (GD) is run on a small sampled subset of clients multiple times before uploading the…
There has been a growing interest in developing data-driven, in particular deep neural network (DNN) based methods for modern communication tasks. For a few popular tasks such as power control, beamforming, and MIMO detection, these methods achieve state-of-the-art performance while requiring less c…
Motivated by large-scale signal processing and machine learning applications, this paper considers the distributed multi-agent optimization problem for a linearly constrained non-convex problem. Each of the agents owns a local cost function and local variable, but are coupled with each other due to…
The non-negative matrix factorization (NMF) model with an additional orthogonality constraint on one of the factor matrices, called the orthogonal NMF (ONMF), has been found to provide improved clustering performance over the K-means. The ONMF model is a challenging optimization problem due to the o…
Identification of cell subclasses using single-cell RNA-Sequencing (scRNA-Seq) data is of paramount importance since it uncovers the hidden biological processes within the cell population. While the nonnegative matrix factorization (NMF) model has been reported to be effective in various unsupervise…
To support multiple on-demand services over several fixed communication networks, the network operators must allow flexible customization and fast provision of their network resources. One effective approach is network virtualization, whereby each service is mapped to a virtual subnetwork providing…
In this paper, we consider a wireless network with one full-duplex (FD) base station (BS) and a set of half-duplex (HD) user equipments (UEs). In such scenario, in addition to the self-interference, the co-channel interference from uplink UEs to downlink UEs is the main bottleneck for the network pe…
Alternating direction method of multipliers (ADMM) has been recognized as an efficient approach for solving many large-scale learning problems over a computer cluster. However, traditional synchronized computation does not scale well with the problem size, as the speed of the algorithm is limited by…
The nonnegative matrix factorization (NMF) has been a popular model for a wide range of signal processing and machine learning problems. It is usually formulated as a nonconvex cost minimization problem. This work settles the convergence issue of a popular algorithm based on the alternating directio…
We consider solving a convex, nonsmooth and stochastic optimization problem over a multi-agent network. Each agent has access to a local objective function and can communicate with its immediate neighbors only. We develop a dynamic stochastic proximal-gradient consensus (DySPGC) algorithm, featuring…
In handling massive-scale signal processing problems arising from `big-data' applications, key technologies could come from the development of decentralized algorithms. In this context, consensus-based methods have been advocated because of their simplicity, fault tolerance and versatility. This pap…
Recently, the alternating direction method of multipliers (ADMM) has been used for distributed consensus optimization and is shown to converge faster than conventional approaches based on consensus subgradient. In this paper, we consider a convex optimization problem with a linearly coupled equality…