Interactive Natural Language Processing (iNLP) has emerged as a novel paradigm within the field of NLP, aimed at addressing limitations in existing frameworks while aligning with the ultimate goals of artificial intelligence. This paradigm considers language models as agents capable of observing, acting, and receiving feedback iteratively from external entities. Specifically, language models in this context can: (1) interact with humans for better understanding and addressing user needs, personalizing responses, aligning with human values, and improving the overall user experience; (2) interact with knowledge bases for enriching language representations with factual knowledge, enhancing the contextual relevance of responses, and dynamically leveraging external information to generate more accurate and informed responses; (3) interact with models and tools for effectively decomposing and addressing complex tasks, leveraging specialized expertise for specific subtasks, and fostering the simulation of social behaviors; and (4) interact with environments for learning grounded representations of language, and effectively tackling embodied tasks such as reasoning, planning, and decision-making in response to environmental observations. This paper offers a comprehensive survey of iNLP, starting by proposing a unified definition and framework of the concept. We then provide a systematic classification of iNLP, dissecting its various components, including interactive objects, interaction interfaces, and interaction methods. We proceed to delve into the evaluation methodologies used in the field, explore its diverse applications, scrutinize its ethical and safety issues, and discuss prospective research directions. This survey serves as an entry point for researchers who are interested in this rapidly evolving area and offers a broad view of the current landscape and future trajectory of iNLP.

Urban areas, as the primary human habitat in modern civilization, accommodate a broad spectrum of social activities. With the surge of embodied intelligence, recent years have witnessed an increasing presence of physical agents in urban areas, such as autonomous vehicles and delivery robots. As a result, practitioners significantly value crafting authentic, simulation-ready 3D cities to facilitate the training and verification of such agents. However, this task is quite challenging. Current generative methods fall short in either diversity, controllability, or fidelity. In this work, we resort to the procedural content generation (PCG) technique for high-fidelity generation. It assembles superior assets according to empirical rules, ultimately leading to industrial-grade outcomes. To ensure diverse and self contained creation, we design a management protocol to accommodate extensive PCG plugins with distinct functions and interfaces. Based on this unified PCG library, we develop a multi-agent framework to transform multi-modal instructions, including OSM, semantic maps, and satellite images, into executable programs. The programs coordinate relevant plugins to construct the 3D city consistent with the control condition. A visual feedback scheme is introduced to further refine the initial outcomes. Our method, named CityX, demonstrates its superiority in creating diverse, controllable, and realistic 3D urban scenes. The synthetic scenes can be seamlessly deployed as a real-time simulator and an infinite data generator for embodied intelligence research. Our project page: this https URL

Graph, such as citation networks, social networks, and transportation networks, are prevalent in the real world. Graph Neural Networks (GNNs) have gained widespread attention for their robust expressiveness and exceptional performance in various graph applications. However, the efficacy of GNNs is heavily reliant on sufficient data labels and complex network models, with the former obtaining hardly and the latter computing costly. To address the labeled data scarcity and high complexity of GNNs, Knowledge Distillation (KD) has been introduced to enhance existing GNNs. This technique involves transferring the soft-label supervision of the large teacher model to the small student model while maintaining prediction performance. This survey offers a comprehensive overview of Graph-based Knowledge Distillation methods, systematically categorizing and summarizing them while discussing their limitations and future directions. This paper first introduces the background of graph and KD. It then provides a comprehensive summary of three types of Graph-based Knowledge Distillation methods, namely Graph-based Knowledge Distillation for deep neural networks (DKD), Graph-based Knowledge Distillation for GNNs (GKD), and Self-Knowledge Distillation based Graph-based Knowledge Distillation (SKD). Each type is further divided into knowledge distillation methods based on the output layer, middle layer, and constructed graph. Subsequently, various algorithms' ideas are analyzed and compared, concluding with the advantages and disadvantages of each algorithm supported by experimental results. In addition, the applications of graph-based knowledge distillation in CV, NLP, RS, and other fields are listed. Finally, the graph-based knowledge distillation is summarized and prospectively discussed. We have also released related resources at this https URL.

Graph Neural Networks (GNNs) are widely adopted for fault diagnosis in microservice systems, premised on their ability to model service dependencies. However, the necessity of explicit graph structures remains underexamined, as existing evaluations conflate preprocessing with architectural contributions. To isolate the true value of GNNs, we propose DiagMLP, a deliberately minimal, topology-agnostic baseline that retains multimodal fusion capabilities while excluding graph modeling. Through ablation experiments across five datasets, DiagMLP achieves performance parity with state-of-the-art GNN-based methods in fault detection, localization, and classification. These findings challenge the prevailing assumption that graph structures are indispensable, revealing that: (i) preprocessing pipelines already encode critical dependency information, and (ii) GNN modules contribute marginally beyond multimodality fusion. Our work advocates for systematic re-evaluation of architectural complexity and highlights the need for standardized baseline protocols to validate model innovations.

With the thriving of pre-trained language model (PLM) widely verified in various of NLP tasks, pioneer efforts attempt to explore the possible cooperation of the general textual information in PLM with the personalized behavioral information in user historical behavior sequences to enhance sequential recommendation (SR). However, despite the commonalities of input format and task goal, there are huge gaps between the behavioral and textual information, which obstruct thoroughly modeling SR as language modeling via PLM. To bridge the gap, we propose a novel Unified pre-trained language model enhanced sequential recommendation (UPSR), aiming to build a unified pre-trained recommendation model for multi-domain recommendation tasks. We formally design five key indicators, namely naturalness, domain consistency, informativeness, noise & ambiguity, and text length, to guide the text-item adaptation and behavior sequence-text sequence adaptation differently for pre-training and fine-tuning stages, which are essential but under-explored by previous works. In experiments, we conduct extensive evaluations on seven datasets with both tuning and zero-shot settings and achieve the overall best performance. Comprehensive model analyses also provide valuable insights for behavior modeling via PLM, shedding light on large pre-trained recommendation models. The source codes will be released in the future.

A simple experiment was conducted to test the ability of the Chinese-based generative artificial intelligence (AI) platform, Wenxin Yige, to render images of urban street views of different countries. The study found that images generated by this AI platform may contain continental-level stereotypes in terms of showing the level of economic development and modernization. Street view images generated from Wenxin Yige do not adequately represent the diverse range of urban landscapes found across different nations. Using these generated images for geography education or outreach initiatives could inadvertently strengthen people's existing stereotypical views about individual countries.

Earthquake forecasting remains a significant scientific challenge, with current methods falling short of achieving the performance necessary for meaningful societal benefits. Traditional models, primarily based on past seismicity and geomechanical data, struggle to capture the complexity of seismic patterns and often overlook valuable non-seismic precursors such as geophysical, geochemical, and atmospheric anomalies. The integration of such diverse data sources into forecasting models, combined with advancements in AI technologies, offers a promising path forward. AI methods, particularly deep learning, excel at processing complex, large-scale datasets, identifying subtle patterns, and handling multidimensional relationships, making them well-suited for overcoming the limitations of conventional approaches. This review highlights the importance of combining AI with geophysical knowledge to create robust, physics-informed forecasting models. It explores current AI methods, input data types, loss functions, and practical considerations for model development, offering guidance to both geophysicists and AI researchers. While many AI-based studies oversimplify earthquake prediction, neglecting critical features such as data imbalance and spatio-temporal clustering, the integration of specialized geophysical insights into AI models can address these shortcomings. We emphasize the importance of interdisciplinary collaboration, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology. By bridging these disciplines, we can develop more accurate, reliable, and societally impactful earthquake forecasting tools.

Physics-informed neural networks (PINNs) have emerged as a transformative framework for addressing operator learning and inverse problems involving the Korteweg-de Vries (KdV) equation for internal solitary waves. By integrating physical constraints with data-driven optimization, PINNs overcome the critical challenges of parameter unmeasurability in the KdV equation for internal solitary waves in two-layer fluid systems. This work addresses two problems: (1) Operator learning constructs a mapping from parameters to solutions, enabling wave evolution predictions from unknown parameters. Comparative studies demonstrate prediction errors as low as $10^{-4}$ when using 1000 training points. (2) Inverse problem solving leverages sparse and potentially noisy observational data with physics-regularized constraints to invert nonlinear coefficients successfully. Compared to conventional approaches, this end-to-end differentiable paradigm unifies operator learning and inverse problem-solving while overcoming mesh discretization errors and high-dimensional parameter space iteration costs. The method shows effectiveness for internal wave problems in stratified fluids, providing both accurate forward modeling and robust parameter inversion capabilities, even under noise.

Optical tactile sensors are extensively utilized in intelligent robot manipulation due to their ability to acquire high-resolution tactile information at a lower cost. However, achieving adequate reality and versatility in simulating optical tactile sensors is challenging. In this paper, we propose a simulation method and validate its effectiveness through experiments. We utilize path tracing for image rendering, achieving higher similarity to real data than the baseline method in simulating pressing scenarios. Additionally, we apply the improved Material Point Method(IMPM) algorithm to simulate the relative rest between the object and the elastomer surface when the object is in motion, enabling more accurate simulation of complex manipulations such as slip and rotation.

Using $7.33~\mathrm{fb}^{-1}$ of $e^+e^-$ collision data taken with the BESIII detector at the BEPCII collider, we report the first experimental study of the purely leptonic decay $D_s^{*+}\to e^+\nu_e$. A signal for the decay $D_s^{*+}\to e^+\nu_e$ is observed with a statistical significance of $2.9\sigma$. The branching fraction of ${D_s^{*+}\to e^+\nu_e}$ is measured to be $(2.1{^{+1.2}_{-0.9}}_{\rm stat.}\pm0.2_{\rm syst.})\times 10^{-5}$, corresponding to an upper limit of $4.0\times10^{-5}$ at the 90\% confidence level. Taking the total width of the $D_s^{*+}$~(($0.070\pm0.028$) keV) predicted by lattice quantum chromodynamics as input, the decay constant of the $D^{*+}_s$ is determined to be $f_{D_s^{*+}}=(213.6{^{+61.0}_{-45.8}}_{\rm stat.}\pm43.9_{\rm syst.})$ MeV, corresponding to an upper limit of 353.8 MeV at the 90\% confidence level.

Using the data sample of an integrated luminosity of 2.93 fb$^{-1}$ taken at the center-of-mass energy of 3.773 GeV, we search for the Majorana neutrino in the lepton number violating decays $D\to K \pi e^+ e^+$. No significant signal is observed, and the upper limits on the branching fraction at the 90\% confidence level are set to be \mathcal{B}\,(D^0 \to K^- \pi^- e^+ e^+)<2.8\times10^{-6}, \mathcal{B}\,(D^+ \to K_S^0 \pi^- e^+ e^+)<3.3\times10^{-6} and \mathcal{B}\,(D^+ \to K^- \pi^0 e^+ e^+)<8.5\times10^{-6}. The Majorana neutrino is searched for with different mass assumptions ranging from 0.25 to 1.0 GeV/$c^2$ in the decays $D^0\to K^- e^+ \nu_N(\pi^- e^+)$ and $D^+\to K_S^0 e^+ \nu_N(\pi^- e^+)$, and the upper limits on the branching fraction at the 90\% confidence level are extracted to be at the level of $10^{-7} \sim 10^{-6}$, depending on the mass of Majorana neutrino. The constraints on the mixing matrix element $|V_{eN}|^2$ are also evaluated.

Small Earth data are geoscience observations with limited short-term monitoring variability, providing sparse but meaningful measurements, typically exhibiting spatiotemporal correlations. Spatiotemporal forecasting on such data is crucial for understanding geoscientific processes despite their small scale. However, conventional deep learning models for spatiotemporal forecasting requires task-specific training for different scenarios. Foundation models do not need task-specific training, but they often exhibit forecasting bias toward the global mean of the pretraining distribution. Here we propose a simple and robust approach for spatiotemporally correlated small Earth data forecasting. The essential idea is to characterize and quantify spatiotemporal patterns of small Earth data and then utilize tabular foundation models for accurate forecasting across different scenarios. Comparative results across three typical scenarios demonstrate that our forecasting approach achieves superior accuracy compared to the graph deep learning model (T-GCN) and tabular foundation model (TabPFN) in the majority of instances, exhibiting stronger robustness.

We present a study of the process $e^+e^-\to\pi^{\pm}(D\bar{D}^*)^{\mp}$ using data samples of 1092~pb$^{-1}$ at $\sqrt{s}=4.23$~GeV and 826~pb$^{-1}$ at $\sqrt{s}=4.26$~GeV collected with the BESIII detector at the BEPCII storage ring. With full reconstruction of the $D$ meson pair and the bachelor $\pi^{\pm}$ in the final state, we confirm the existence of the charged structure $Z_c(3885)^{\mp}$ in the $(D\bar{D}^*)^{\mp}$ system in the two isospin processes $e^+e^-\to\pi^+D^0D^{*-}$ and $e^+e^-\to\pi^+D^-D^{*0}$. By performing a simultaneous fit, the statistical significance of $Zc(3885)^{\mp}$ signal is determined to be greater than 10$\sigma$, and its pole mass and width are measured to be $M_{\rm{pole}}$=(3881.7$\pm$1.6(stat.)$\pm$1.6(syst.))~MeV/$c^2$ and $\Gamma_{\rm{pole}}$=(26.6$\pm$2.0(stat.)$\pm$2.1(syst.))~MeV, respectively. The Born cross section times the $(D\bar{D}^*)^{\mp}$ branching fraction ($\sigma(e^+e^-\to\pi^{\pm}Z_{c}(3885)^{\mp}) \times Br(Z_{c}(3885)^{\mp}\to(D\bar{D}^*)^{\mp})$) is measured to be $(141.6\pm7.9(\text{stat.})\pm12.3(\text{syst.}))~\text{pb}$ at $\sqrt{s}=4.23$~GeV and $(108.4\pm6.9(\text{stat.})\pm8.8(\text{syst.}))~\text{pb}$ at $\sqrt{s}=4.26$~GeV. The polar angular distribution of the $\pi^{\pm}$-$Z_c(3885)^{\mp}$ system is consistent with the expectation of a quantum number assignment of $J^P=1^+$ for $Z_c(3885)^{\mp}$.

Based on $(10087\pm44)\times10^{6}$ $J/\psi$ events collected with the BESIII detector, a partial wave analysis of the decay $J/\psi\rightarrow\gamma K^{0}_{S}K^{0}_{S}\eta^{\prime}$ is performed. The mass and width of the $X(2370)$ are measured to be $2395 \pm 11 ({\rm stat})^{+26}_{-94}({\rm syst})\ \mathrm{MeV}/c^{2}$and$188^{+18}_{-17}({\rm stat})^{+124}_{-33}({\rm syst})~\mathrm{MeV}$, respectively. The corresponding product branching fraction is $\mathcal{B}[J/\psi\rightarrow\gamma X(2370)] \times \mathcal{B}[X(2370) \rightarrow f_{0}(980)\eta^{\prime}] \times \mathcal{B}[f_{0}(980) \rightarrow K^{0}_{S}K^{0}_{S}] = \left( 1.31 \pm 0.22 ({\rm stat})^{+2.85}_{-0.84}({\rm syst}) \right) \times 10^{-5}$. The statistical significance of the $X(2370)$ is greater than $11.7\sigma$ and the spin-parity is determined to be $0^{-+}$ for the first time. The measured mass and spin-parity of the $X(2370)$ are consistent with the predictions of the lightest pseudoscalar glueball.

A search has been performed for the semileptonic decays $D^{0}\to K_{S}^{0} K^{-} e^{+}\nu_{e}$, $D^{+}\to K_{S}^{0} K_{S}^{0} e^{+}\nu_{e}$ and $D^{+}\to K^{+}K^{-} e^{+}\nu_{e}$, using $7.9~\mathrm{fb}^{-1}$ of $e^+e^-$ annihilation data collected at the center-of-mass energy $\sqrt{s}=3.773$ GeV by the BESIII detector operating at the BEPCII collider. No significant signals are observed, and upper limits are set at the 90\% confidence level of $2.13\times10^{-5}$, $1.54\times10^{-5}$ and $2.10\times10^{-5}$ for the branching fractions of $D^{0}\to K_{S}^{0} K^{-} e^{+}\nu_{e}$, $D^{+}\to K_{S}^{0} K_{S}^{0} e^{+}\nu_{e}$ and $D^{+}\to K^{+}K^{-} e^{+}\nu_{e}$, respectively.