Abstract Existing LLM-based role-playing methods often rely on superficial textual descriptions or simplistic metrics, inadequately modeling both intrinsic and extrinsic character dimensions. Additionally, they typically simulate character memory with implicit model knowledge or basic retrieval augment generation without explicit memory alignment, compromising memory consistency. The two issues weaken reliability of role-playing LLMs in several applications, such as trustworthy social simulation. To address these limitations, we propose PsyMem, a novel framework integrating fine-grained psychological attributes and explicit memory control for role-playing. PsyMem supplements textual descriptions with 26 psychological indicators to detailed model character. Additionally, PsyMem implements memory alignment training, explicitly trains the model to align character’s response with memory, thereby enabling dynamic memory-controlled responding during inference. By training Qwen2.5-7B-Instruct on our specially designed dataset (including 5,414 characters and 38,962 dialogues extracted from novels), the resulting model, termed as PsyMem-Qwen, outperforms baseline models in role-playing, achieving the best performance in human-likeness and character fidelity.

Nonlinear probabilistic latent variable models (NPLVMs) are a cornerstone of soft sensor modeling due to their capacity for uncertainty delineation. However, conventional NPLVMs are trained using amortized variational inference, where neural networks parameterize the variational posterior. While facilitating model implementation, this parameterization converts the distributional optimization problem within an infinite-dimensional function space to parameter optimization within a finite-dimensional parameter space, which introduces an approximation error gap, thereby degrading soft sensor modeling accuracy. To alleviate this issue, we introduce KProxNPLVM, a novel NPLVM that pivots to relaxing the objective itself and improving the NPLVM's performance. Specifically, we first prove the approximation error induced by the conventional approach. Based on this, we design the Wasserstein distance as the proximal operator to relax the learning objective, yielding a new variational inference strategy derived from solving this relaxed optimization problem. Based on this foundation, we provide a rigorous derivation of KProxNPLVM's optimization implementation, prove the convergence of our algorithm can finally sidestep the approximation error, and propose the KProxNPLVM by summarizing the abovementioned content. Finally, extensive experiments on synthetic and real-world industrial datasets are conducted to demonstrate the efficacy of the proposed KProxNPLVM.

Missing data imputation through distribution alignment has demonstrated advantages for non-temporal datasets but exhibits suboptimal performance in time-series applications. The primary obstacle is crafting a discrepancy measure that simultaneously (1) captures temporal patterns—accounting for periodicity and temporal dependencies inherent in time-series—and (2) accommodates non-stationarity, ensuring robustness amidst multiple coexisting temporal patterns. In response to these challenges, we introduce the Proximal Spectrum Wasserstein (PSW) discrepancy, a novel discrepancy tailored for comparing two \textit{sets} of time-series based on optimal transport. It incorporates a pairwise spectral distance to encapsulate temporal patterns, and a selective matching regularization to accommodate non-stationarity. Subsequently, we develop the PSW for Imputation (PSW-I) framework, which iteratively refines imputation results by minimizing the PSW discrepancy. Extensive experiments demonstrate that PSW-I effectively accommodates temporal patterns and non-stationarity, outperforming prevailing time-series imputation methods. Code is available at https://github.com/FMLYD/PSW-I.

Iterative imputation is a prevalent method for missing data imputation, where each feature is imputed iteratively by treating it as a target variable estimated from all other features. However, iterative imputation method suffers from two principal limitations: (1) it imposes a single parametric model form to impute all features, neglecting the potential for optimal models to vary among features, which risks model misspecification; and (2) it assumes every feature contains missing values, overlooking the potential presence of non-missing features, termed as oracle features, which are informative for imputation. To address these limitations, we propose kernel point imputation (KPI), a bi-level optimization framework for iterative missing data imputation. At the inner level, KPI adaptively learns the optimal model form for each feature within a reproducing kernel Hilbert space, addressing limitation (1). At the outer level, KPI utilizes oracle features as supervisory signals to iteratively refine the imputations, addressing limitation (2). Experiments demonstrate that KPI outperforms competitive imputation methods. Code is available at https://github.com/FMLYD/kpi.git.

Time series data can be represented in both the time and frequency domains, with the time domain emphasizing local dependencies and the frequency domain highlighting global dependencies. To harness the strengths of both domains in capturing local and global dependencies, we propose a novel Frequency and Time Domain Mixer (FTMixer) method. To exploit the global characteristics of the frequency domain, we introduce a novel Frequency Channel Convolution (FCC) module, designed to capture global inter-series dependencies. Inspired by the windowing concept in frequency domain transformations, we further propose a novel Windowed Frequency-Time Convolution (WFTC) module, which captures local dependencies by leveraging both frequency domain representations obtained from windowed transformations and time domain representations. Notably, FTMixer employs the Discrete Cosine Transformation (DCT) with real numbers instead of the complex-number-based Discrete Fourier Transformation (DFT), enabling direct utilization of modern deep learning operators in the frequency domain. Extensive experimental results across seven real-world long-term time series datasets demonstrate the superiority of FTMixer, in terms of both forecasting performance and computational efficiency.

Recommendation systems are pivotal in improving user experience on various digital platforms. However, observational training data in recommendation systems introduce selection bias, which leads to a distributional discrepancy between training data and real-world scenarios, resulting in suboptimal performance. Current causal debiasing methods such as inverse propensity score and doubly robust rely on accurately estimated propensity scores, typically optimized through negative log-likelihood (NLL) minimization. However, recent studies have highlighted the limitations of this approach, as perfect NLL minimization may not adequately correct for selection bias. To address this issue, we propose Wasserstein Balancing Metric (WBM), a novel metric that measures and enhances the balancing capacity of propensity scores in causal debiasing methods by minimizing the Wasserstein discrepancy between reweighted populations. On the basis, we introduce IPS-WBM and DR-WBM, incorporating WBM as a regularizer in standard inverse propensity score and doubly robust estimators, which enhances causal balancing capacity without introducing additional bias. Extensive experiments on three real-world recommendation datasets demonstrate that our methods improve the causal balancing capability of learned propensities and enhance debiasing performance.

We present an investigation into the design of an acoustic communication network, where multiple users are distributed across space and transmit and receive simultaneously in the same band to and from a common base station. Specifically, we focus on a system that utilizes orthogonal frequency division multiplexing as a modulation method, and allows users to transmit and receive in either synchronous or asynchronous fashion. To distinguish between users on the uplink, the base station employs a combination of code-division and space-division multiple access. The base station iteratively steers a beam to each stable propagation path of the desired user’s channel while placing nulls in the direction of other paths, as well as in the directions of interfering users. Finally, the multiple paths of the desired user are recombined before data detection. The process is repeated for each user. Broadband beamforming is employed to account for the broadband nature of acoustic signals. The beamformer coefficients on each carrier depend on the angles of signal arrivals, which are estimated during the uplink transmission and used to construct both the uplink and the downlink beamformer. On the downlink, the base station utilizes angle information to assemble beamforming weights and point in the direction of stable paths of the users. It superimposes multiuser signals and transmits the sum signal to the users. The signal intended for a given user reaches only that user, requiring just a simple detector. Each user is equipped by a single-element transducer. To demonstrate the design concepts, we conducted simulations using a shallow water channel model and performed experimental over-the-air tests in an indoor environment using an acoustic communications testbed. The results were excellent, thereby encouraging future implementations in practical systems.

The growing spectrum crunch has motivated exploratory efforts in the use of LTE in the 5 GHz bands for downlink traffic. However, this paradigm raises concerns of fair sharing of the spectrum and the adverse impact of scheduled LTE frames on Wi-Fi Packet Success Rates (PSR). To address this issue, we propose E-Fi, an interference-evasion mechanism that allows Wi-Fi devices to survive LTE transmissions without any cooperation between these two different standards. Different from existing approaches, we argue that the simple use of Almost Blank Subframes (ABS) within the LTE standard offering short channel access windows overestimates opportunities for Wi-Fi. The pilots embedded in the ABS not only interfere with Wi-Fi but also adversely impact the carrier sensing function. E-Fi mitigates this problem through a two-fold approach. It uses a combination of (i) Wi-Fi Direct with packet relaying and (ii) classical distributed coordination function to reach distant nodes. Second, it ensures load balancing for both Wi-Fi uplink and downlink traffic with high PSR by creating node-groups based with dedicated contention-based medium access intervals. Our approach is validated by comprehensive simulation and experimental results that indicate significantly higher throughput in E-Fi compared to classical Wi-Fi.