Ruiyu Zhang

Ruiyu Zhang, Qihao Wang

Multimodal density estimation is a fundamental problem in scientific computing, but Python has lacked a cohesive implementation of critical bandwidth analysis and related mode-counting tools. We present critband, a Python package for Silverman-style critical bandwidth detection and multimodality assessment based on Gaussian kernel density estimation. The package provides a hybrid binary-Brent critical bandwidth solver, FFT-accelerated KDE for large samples, bootstrap confidence intervals with explicit failure accounting, support for arbitrary mode-count thresholds, component decomposition, bimodality strength scoring, and excess-mass estimation. We validate the implementation on twelve benchmark mixtures spanning well-separated, borderline, unequal-variance, unequal-weight, trimodal, and small-sample cases, and cross-check results against the R multimode package. Across the tested cases, critband reproduces critical bandwidth behavior closely and provides substantially faster runtime for the core search routine. The package is distributed on PyPI and is designed to integrate cleanly into the Python scientific computing stack.

Yuhao Zhang, Zhiheng Liu, Fan Bu et al.

Existing end-to-end speech large language models (LLMs) usually rely on large-scale annotated data for training, while data-efficient training has not been discussed in depth. We focus on two fundamental problems between speech and text: the representation space gap and sequence length inconsistency. We propose Soundwave, which utilizes an efficient training strategy and a novel architecture to address these issues. Results show that Soundwave outperforms the advanced Qwen2-Audio in speech translation and AIR-Bench speech tasks, using only one-fiftieth of the training data. Further analysis shows that Soundwave still retains its intelligence during conversation. The project is available at https://github.com/FreedomIntelligence/Soundwave.

Ruiyu Zhang, Lin Nie, Ce Zhao et al.

Accurately interpreting words is vital in political science text analysis; some tasks require assuming semantic stability, while others aim to trace semantic shifts. Traditional static embeddings, like Word2Vec effectively capture long-term semantic changes but often lack stability in short-term contexts due to embedding fluctuations caused by unbalanced training data. BERT, which features transformer-based architecture and contextual embeddings, offers greater semantic consistency, making it suitable for analyses in which stability is crucial. This study compares Word2Vec and BERT using 20 years of People's Daily articles to evaluate their performance in semantic representations across different timeframes. The results indicate that BERT outperforms Word2Vec in maintaining semantic stability and still recognizes subtle semantic variations. These findings support BERT's use in text analysis tasks that require stability, where semantic changes are not assumed, offering a more reliable foundation than static alternatives.

Ruiyu Zhang, Lin Nie

This study integrates critical AI scholarship with relational communication theories to explain how AI language modifications shape the quality of government-citizen communication. Distinguishing between informational-cognitive quality (clarity, ease of response) and expressive-constitutive quality (politeness, respectfulness, feeling heard, trust, urgency, empathy), we hypothesize that AI yields uncontested benefits for the former but contested effects for the latter, potentially enhancing relational markers while muting authentic emotional cues. Using a vignette-based survey with 220 citizens and 214 civil servants in China, we assess perceptions across five interaction contexts: service requests, policy inquiries, complaints, suggestions, and emergencies. Results from paired t-tests and mixed-effects regressions support the claim that AI enhances both informational-cognitive and expressive-constitutive quality from the perspectives of citizens and civil servants, with significant improvements in clarity, politeness, satisfaction, trust, and empathy, but provide no consistent evidence of urgency or empathy signals. These findings suggest that concerns over algorithmic emotional flattening may be overstated or context-specific; they offer theoretical insights into AI-mediated public interactions and practical implications for fostering trust and efficiency in digital governance.

Ruiyu Zhang, Lin Nie, Xin Zhao

Hierarchical Evolutionary Agent Simulation (HEAS) is a Python framework that unifies layered agent-based modeling with evolutionary optimization and tournament evaluation in a single, reproducible workflow. HEAS represents models as hierarchies of lightweight processes ("streams") scheduled in deterministic layers that read and write a shared context, making cross-scale couplings explicit and auditable. A compact API and CLI-simulate, optimize, evaluate-expose single- and multi-objective evolution, PyTorch policy integration via parameter flattening/unflattening, and general tournament tooling with user-defined scoring and voting rules. The framework standardizes evaluation through uniform per-step and episode metrics, persists seeds, logbooks, and hall-of-fame archives, and provides plotting helpers for traces, Pareto fronts, and comparative outcomes, reducing glue code and improving comparability across studies. HEAS emphasizes separation of mechanism from orchestration, allowing exogenous drivers, endogenous agents, and aggregators to be composed and swapped without refactoring, while the same model can be used for forward simulation, optimization, or systematic comparison. We illustrate usage with two compact examples-an ecological system and an enterprise decision-making setting. HEAS offers a practical foundation for cross-disciplinary, multi-level inquiry, yielding reliable, reproducible results.

Ruiyu Zhang, Ce Zhao, Xin Zhao et al.

Learning interpretable latent representations from tabular data remains a challenge in deep generative modeling. We introduce SE-VAE (Structural Equation-Variational Autoencoder), a novel architecture that embeds measurement structure directly into the design of a variational autoencoder. Inspired by structural equation modeling, SE-VAE aligns latent subspaces with known indicator groupings and introduces a global nuisance latent to isolate construct-specific confounding variation. This modular architecture enables disentanglement through design rather than through statistical regularizers alone. We evaluate SE-VAE on a suite of simulated tabular datasets and benchmark its performance against a series of leading baselines using standard disentanglement metrics. SE-VAE consistently outperforms alternatives in factor recovery, interpretability, and robustness to nuisance variation. Ablation results reveal that architectural structure, rather than regularization strength, is the key driver of performance. SE-VAE offers a principled framework for white-box generative modeling in scientific and social domains where latent constructs are theory-driven and measurement validity is essential.