清华大学

In this article, we propose a novel model for time series of counts called the hysteretic Poisson autoregressive (HPART) model with thresholds by extending the linear Poisson autoregressive model into a nonlinear model. Unlike other approaches that bear the adjective “hysteretic", our model incorporates a scientifically relevant controlling factor that produces genuine hysteresis. Further, we re-analyse the buffered Poisson autoregressive (BPART) model with thresholds. Although the two models share the convenient piecewise linear structure, the HPART model probes deeper into the intricate dynamics that governs regime switching. We study the maximum likelihood estimation of the parameters of both models and their asymptotic properties in a unified manner, establish tests of separate families of hypotheses for the non-nested case involving a BPART model and a HPART model, and demonstrate the finite-sample efficacy of parameter estimation and tests with Monte Carlo simulation. We showcase advantages of the HPART model with two real time series, including plausible interpretations and improved out-of-sample predictions.

Wasserstein autoregression provides a robust framework for modeling serial dependence among probability distributions, with wide-ranging applications in economics, finance, and climate science. In this paper, we develop portmanteau-type diagnostic tests for assessing the adequacy of Wasserstein autoregressive models. By defining autocorrelation functions for model errors and residuals in the Wasserstein space, we construct two related tests: one analogous to the classical McLeod type test, and the other based on the sample-splitting approach of Davis and Fernandes(2025). We establish that, under mild regularity conditions, the corresponding test statistics converge in distribution to chi-square limits. Simulation studies and empirical applications demonstrate that the proposed tests effectively detect model mis-specification, offering a principled and reliable diagnostic tool for distributional time series analysis.

Asking appropriate follow-up questions is an important aspect of semi-structured interviews, yet interviewers’ knowledge and skills may limit their depth and affect outcomes. While many studies have demonstrated the usefulness of large language models (LLMs) for qualitative analysis, their potential in data collection remains underexplored. We adopt an AI-driven/Reverse “Wizard-of-Oz” setup to investigate how real-time LLM support for generating follow-up questions shapes semi-structured interviews. Through a study with 17 participants, we examine the value of LLM-generated follow-up questions, the evolving division of roles, relationships, collaborative behaviors, and responsibilities between interviewers and AI. Our findings (1) provide empirical evidence of the strengths and limitations of AI-generated follow-up questions (AGQs); (2) introduce a Human-AI collaboration framework in this interview context; and (3) propose human-centered design guidelines for AI-assisted interviews. We position LLMs as complements, not replacements, human judgment, and highlight pathways for integrating AI into qualitative data collection.

Vision-language models (VLMs) have demonstrated remarkable capabilities in robotic planning, particularly for long-horizon tasks that require a holistic understanding of the environment for task decomposition. Existing methods typically rely on prior environmental knowledge or carefully designed task-specific prompts, making them struggle with dynamic scene changes or unexpected task conditions. Such challenges underscore two critical issues: adaptability and efficiency. To address them, in this work, we propose an adaptive multi-agent planning framework, termed REMAC, that enables efficient, scene-agnostic multi-robot long-horizon task planning and execution through continuous reflection and self-evolvement. REMAC incorporates two key modules: a self-reflection module performing pre-condition and post-condition checks in the loop to evaluate progress and refine plans, and a self-evolvement module dynamically adapting plans based on scene-specific reasoning. It offers appealing benefits: (1) Robots can keep reflecting on potential planning errors and adapting the plan based on task-specific insights. (2) After iterations, a robot can call another one to coordinate tasks in parallel, maximizing the task execution efficiency. To validate REMAC’s effectiveness, we build a multi-agent environment for long-horizon robot manipulation and navigation based on RoboCasa, featuring 4 task categories with 27 task styles and 50+ different objects. Based on it, we further benchmark state-of-the-art reasoning models, including DeepSeek-R1, o3, QwQ, Qwen3, and Grok3. Extensive experiments demonstrate REMAC’s superiority by boosting average success rates by 40% and execution efficiency by 52.7% over the single robot baseline without any task-specific prompting or fine-tuning in initial planning. Code, demos, and appendix are available at the website https://reamac-repo.github.io/remac-project-page/.

The identification of dq-asymmetric impedances presents a challenging estimation problem due to the strong cross-coupling between axes, which is especially prominent at low frequencies. Conventional identification schemes are often hampered by the requirement for sequential perturbations or the difficulty of selecting appropriate global parametric models. This paper introduces an active non-parametric frequency-domain approach that utilizes complex transfer functions and local parametric approximations. This formulation eliminates the requirement for periodic steady-state measurements while providing precision comparable to that of correctly specified parametric models. By solving localized linear least-squares problems, the approach maintains high computational efficiency while remaining agnostic to the specific grid structure. Electromagnetic transient (EMT) simulations demonstrate that the proposed method achieves high-fidelity identification with a 1 Hz frequency resolution using only one second of noise-corrupted measurements. The results confirm a significant reduction in measurement time compared to available techniques. Additionally, the approach shows superior robustness against model-order selection errors, ensuring accurate wideband modeling of grid dynamics.