论文检索

Deep Learning · Algorithms

Weilin Wan, Weizhong Zhang, Cheng Jin

Data selection improves computational efficiency by choosing informative subsets of training samples. However, existing methods ignore the compute budget, treating data selection and importance evaluation independently of compute budget constraints. Yet empirical studies show no algorithm can consistently outperform others (or even random selection) across varying budgets. We therefore argue that compute budget must be integral to data-selection strategies, since different budgets impose distinct requirements on data quantity, quality, and distribution for effective training. To this end, we propose a novel Computational budget-Aware Data Selection (CADS) method and naturally formulate it into a bilevel optimization framework, where the inner loop trains the model within the constraints of the computational budget on some selected subset of training data, while the outer loop optimizes data selection based on model evaluation. Our technical contributions lie in addressing two main challenges in solving this bilevel optimization problem: the expensive Hessian matrix estimation for outer-loop gradients and the computational burden of achieving inner-loop optimality during iterations. To solve the first issue, we propose a probabilistic reparameterization strategy and compute the gradient using a Hessian-free policy gradient estimator. To address the second challenge, we transform the inner optimization problem into a penalty term in the outer objective, further discovering that we only need to estimate the minimum of a one-dimensional loss to calculate the gradient, significantly improving efficiency. To accommodate different data selection granularities, we present two complementary CADS variants: an example-level version (CADS-E) offering fine-grained control and a source-level version (CADS-S) aggregating samples into source groups for scalable, efficient selection without sacrificing effectiveness. Extensive experiments show that our method achieves performance gains of up to 14.42\% over baselines in vision and language benchmarks. Additionally, CADS achieves a 3-20× speedup compared to conventional bilevel implementations, with acceleration correlating positively with compute budget size.

Theory · Everything Else

Lucas Morisset, Adrien Hardy, Alain Durmus

This paper addresses the problem of inverse covariance (also known as precision matrix) estimation in high-dimensional settings. Specifically, we focus on two classes of estimators: linear shrinkage estimators with a target proportional to the identity matrix, and estimators derived from data augmentation (DA). Here, DA refers to the common practice of enriching a dataset with artificial samples—typically generated via a generative model or through random transformations of the original data—prior to model fitting. For both classes of estimators, we derive estimators and provide concentration bounds for their quadratic error. This allows for both method comparison and hyperparameter tuning, such as selecting the optimal proportion of artificial samples. On the technical side, our analysis relies on tools from random matrix theory. We introduce a novel deterministic equivalent for generalized resolvent matrices, accommodating dependent samples with specific structure. We support our theoretical results with numerical experiments.

Deep Learning · Algorithms

Lancheng Zou, Shuo Yin, Zehua Pei, Tsung-Yi Ho, Farzan Farnia, Bei Yu

Channel permutation is a powerful technique for enhancing the accuracy of N:M sparse models by reordering the channels of weight matrices to prioritize the retention of important weights. However, traditional channel permutation methods rely on handcrafted quality metrics, which often fail to accurately capture the true impact of pruning on model performance. To address this limitation, we propose PermLLM, a novel post-training pruning framework that introduces learnable channel permutation (LCP) for N:M sparsity. LCP leverages Sinkhorn normalization to transform discrete permutation matrices into differentiable soft permutation matrices, enabling end-to-end optimization. Additionally, PermLLM incorporates an efficient block-wise channel permutation strategy, which significantly reduces the number of learnable parameters and computational complexity. PermLLM seamlessly integrates with existing one-shot pruning methods to adaptively optimize channel permutations, effectively mitigating pruning-induced errors. Extensive experiments on the LLaMA series, Qwen, and OPT models demonstrate that PermLLM achieves superior performance in optimizing N:M sparse models.

Social Aspects · Trustworthy Machine Learning

Chen Xiong, Pin-Yu Chen, Tsung-Yi Ho

Recent advances in Large Language Models (LLMs) have spurred transformative applications in various domains, ranging from open-source to proprietary LLMs. However, jailbreak attacks, which aim to break safety alignment and user compliance by tricking the target LLMs into answering harmful and risky responses, are becoming an urgent concern. The practice of red-teaming for LLMs is to proactively explore potential risks and error-prone instances before the release of frontier AI technology. This paper proposes an agentic workflow to automate and scale the red-teaming process of LLMs through the Composition-of-Principles (CoP) framework, where human users provide a set of red-teaming principles as instructions to an AI agent to automatically orchestrate effective red-teaming strategies and generate jailbreak prompts. Distinct from existing red-teaming methods, our CoP framework provides a unified and extensible framework to encompass and orchestrate human-provided red-teaming principles to enable the automated discovery of new red-teaming strategies. When tested against leading LLMs, CoP reveals unprecedented safety risks by finding novel jailbreak prompts and improving the best-known single-turn attack success rate by up to 13.8 times.

Computer Vision · Classification and Understanding

Wenhao Li, Qiangchang Wang, Xianjing Meng, Zhibin Wu, Yilong Yin

Few-shot learning (FSL) aims to recognize novel concepts from only a few labeled support samples. Recent studies enhance support features by incorporating additional semantic information (e.g., class descriptions) or designing complex semantic fusion modules. However, these methods still suffer from hallucinating semantics that contradict the visual evidence due to the lack of grounding in actual instances, resulting in noisy guidance and costly corrections. To address these issues, we propose a novel framework, bridging Vision and Text with LLMs for Few-Shot Learning (VT-FSL), which constructs precise cross-modal prompts conditioned on Large Language Models (LLMs) and support images, seamlessly integrating them through a geometry-aware alignment mechanism. It mainly consists of Cross-modal Iterative Prompting (CIP) and Cross-modal Geometric Alignment (CGA). Specifically, the CIP conditions an LLM on both class names and support images to generate precise class descriptions iteratively in a single structured reasoning pass. These descriptions not only enrich the semantic understanding of novel classes but also enable the zero-shot synthesis of semantically consistent images. The descriptions and synthetic images act respectively as complementary textual and visual prompts, providing high-level class semantics and low-level intra-class diversity to compensate for limited support data. Furthermore, the CGA jointly aligns the fused textual, support, and synthetic visual representations by minimizing the kernelized volume of the 3-dimensional parallelotope they span. It captures global and nonlinear relationships among all representations, enabling structured and consistent multimodal integration. The proposed VT-FSL method establishes new state-of-the-art performance across ten diverse benchmarks, including standard, cross-domain, and fine-grained few-shot learning scenarios. Code is available at https://github.com/peacelwh/VT-FSL.

Deep Learning · Everything Else

Weize Chen, Jiarui yuan, Jin Tailin, Ning Ding, Huimin Chen, Zhiyuan Liu, Maosong Sun

Recent large language models (LLMs) exhibit impressive reasoning but often \textit{overthink}, generating excessively long responses that hinder efficiency. We introduce DIET (DIfficulty-AwarE Training), a framework that systematically cuts these "token calories" by integrating on-the-fly problem difficulty into the reinforcement learning (RL) process. DIET dynamically adapts token compression strategies by modulating token penalty strength and conditioning target lengths on estimated task difficulty, to optimize the performance-efficiency trade-off. We also theoretically analyze the pitfalls of naive reward weighting in group-normalized RL algorithms like GRPO, and propose \textit{Advantage Weighting} technique, which enables stable and effective implementation of these difficulty-aware objectives. Experimental results demonstrate that DIET significantly reduces token counts while simultaneously improving reasoning performance. Beyond raw token reduction, we show two crucial benefits largely overlooked by prior work: (1) DIET leads to superior \textbf{inference scaling}. By maintaining high per-sample quality with fewer tokens, it enables better scaling performance via majority voting under fixed computational budgets, an area where other methods falter. (2) DIET enhances the natural positive correlation between response length and problem difficulty, ensuring verbosity is appropriately allocated, unlike many existing compression methods that disrupt this relationship. Our analyses provide a principled and effective framework for developing more efficient, practical, and high-performing LLMs.

Data-centric AI · Everything Else

Dominic Broadbent, Nick Whiteley, Robert Allison, Tom Lovett

Existing distribution compression methods, like Kernel Herding (KH), were originally developed for unlabelled data. However, no existing approach directly compresses the conditional distribution of *labelled* data. To address this gap, we first introduce the *Average Maximum Conditional Mean Discrepancy* (AMCMD), a metric for comparing conditional distributions, and derive a closed form estimator. Next, we make a key observation: in the context of distribution compression, the cost of constructing a compressed set targeting the AMCMD can be reduced from $\mathcal{O}(n^3)$ to $\mathcal{O}(n)$. Leveraging this, we extend KH to propose Average Conditional Kernel Herding (ACKH), a linear-time greedy algorithm for constructing compressed sets that target the AMCMD. To better understand the advantages of *directly* compressing the conditional distribution rather than doing so via the joint distribution, we introduce *Joint Kernel Herding* (JKH), an adaptation of KH designed to compress the joint distribution of labelled data. While herding methods provide a simple and interpretable selection process, they rely on a greedy heuristic. To explore alternative optimisation strategies, we also propose *Joint Kernel Inducing Points* (JKIP) and *Average Conditional Kernel Inducing Points* (ACKIP), which *jointly* optimise the compressed set while maintaining linear complexity. Experiments show that directly preserving conditional distributions with ACKIP outperforms both joint distribution compression and the greedy selection used in ACKH. Moreover, we see that JKIP consistently outperforms JKH.

Theory · Everything Else

Dhruva Karkada, James Simon, Yasaman Bahri, Michael Deweese

Self-supervised word embedding algorithms such as word2vec provide a minimal setting for studying representation learning in language modeling. We examine the quartic Taylor approximation of the word2vec loss around the origin, and we show that both the resulting training dynamics and the final performance on downstream tasks are empirically very similar to those of word2vec. Our main contribution is to analytically solve for both the gradient flow training dynamics and the final word embeddings in terms of only the corpus statistics and training hyperparameters. The solutions reveal that these models learn orthogonal linear subspaces one at a time, each one incrementing the effective rank of the embeddings until model capacity is saturated. Training on Wikipedia, we find that each of the top linear subspaces represents an interpretable topic-level concept. Finally, we apply our theory to describe how linear representations of more abstract semantic concepts emerge during training; these can be used to complete analogies via vector addition.

Theory · Everything Else

Yaroslav Mukhin

The influence function (IF) of a statistical functional is the Riesz representer of its derivative, also known as its first variation and Fisher-Rao gradient. It is a key object for numerical optimization over probability measures, semiparametric efficiency theory, standard constructions of efficient estimators, and an arsenal of inference methods for these estimators. Yet, deriving the IF analytically is often an obstruction for practitioners. To automate this task, we develop a novel spectral representation of the IF that lends itself to a low-rank functional estimator in a reproducing kernel Hilbert space (rkHs). Our estimator (i) does not require analytic derivations by the user, (ii) relies on kernel Principal Component Analysis and numerical pathwise derivatives along these components. We present the derivation of the representation and prove consistency of the low-rank rkHs estimator.

Deep Learning · Generative Models and Autoencoders

Paul Krzakala, Gabriel Melo, Charlotte Laclau, Florence d'Alché-Buc, Rémi Flamary

Although graph-based learning has attracted a lot of attention, graph representation learning is still a challenging task whose resolution may impact key application fields such as chemistry or biology. To this end, we introduce GRALE, a novel graph autoencoder that encodes and decodes graphs of varying sizes into a shared embedding space. GRALE is trained using an Optimal Transport-inspired loss that compares the source and reconstructed graphs and leverages a differentiable matching module, which is trained jointly with the encoder and decoder. The proposed attention-based architecture relies on Evoformer, the core component of AlphaFold, which we extend to support both graph encoding and decoding. We show, in numerical experiments on simulated and molecular data, that GRALE enables a highly general form of pre-training, applicable to a wide range of downstream tasks, from classification and regression to more complex tasks such as graph interpolation, editing, matching, and prediction.

Computer Vision · Everything Else

Jinghong Zheng, Changlong Jiang, Yang Xiao, Jiaqi Li, Haohong Kuang, Hang Xu, Ran Wang, Zhiguo Cao, Min Du, Joey Tianyi Zhou

3D human pose lifting from a single RGB image is a challenging task in 3D vision. Existing methods typically establish a direct joint-to-joint mapping from 2D to 3D poses based on 2D features. This formulation suffers from two fundamental limitations: inevitable error propagation from input predicted 2D pose to 3D predictions and inherent difficulties in handling self-occlusion cases. In this paper, we propose PandaPose, a 3D human pose lifting approach via propagating 2D pose prior to 3D anchor space as the unified intermediate representation. Specifically, our 3D anchor space comprises: (1) Joint-wise 3D anchors in the canonical coordinate system, providing accurate and robust priors to mitigate 2D pose estimation inaccuracies. (2) Depth-aware joint-wise feature lifting that hierarchically integrates depth information to resolve self-occlusion ambiguities. (3) The anchor-feature interaction decoder that incorporates 3D anchors with lifted features to generate unified anchor queries encapsulating joint-wise 3D anchor set, visual cues and geometric depth information. The anchor queries are further employed to facilitate anchor-to-joint ensemble prediction. Experiments on three well-established benchmarks (i.e., Human3.6M, MPI-INF-3DHP and 3DPW) demonstrate the superiority of our proposition. The substantial reduction in error by 14.7% compared to SOTA methods on the challenging conditions of Human3.6M and qualitative comparisons further showcase the effectiveness and robustness of our approach.

Computer Vision · Everything Else

Qiyuan Guan, Xiang Chen, Guiyue Jin, Jiyu Jin, Shumin Fan, Tianyu Song, Jinshan Pan

Compared to daytime image deraining, nighttime image deraining poses significant challenges due to inherent complexities of nighttime scenarios and the lack of high-quality datasets that accurately represent the coupling effect between rain and illumination. In this paper, we rethink the task of nighttime image deraining and contribute a new high-quality benchmark, HQ-NightRain, which offers higher harmony and realism compared to existing datasets. In addition, we develop an effective Color Space Transformation Network (CST-Net) for better removing complex rain from nighttime scenes. Specifically, we propose a learnable color space converter (CSC) to better facilitate rain removal in the Y channel, as nighttime rain is more pronounced in the Y channel compared to the RGB color space. To capture illumination information for guiding nighttime deraining, implicit illumination guidance is introduced enabling the learned features to improve the model's robustness in complex scenarios. Extensive experiments show the value of our dataset and the effectiveness of our method. The source code and datasets are available at https://github.com/guanqiyuan/CST-Net.

Reinforcement Learning · Multi-agent

Fivos Kalogiannis, Gabriele Farina

Multi-agent reinforcement learning (MARL) has long been seen as inseparable from Markov games (Littman 1994). Yet, the most remarkable achievements of practical MARL have arguably been in extensive-form games (EFGs)---spanning games like Poker, Stratego, and Hanabi. At the same time, little is known about provable equilibrium convergence for MARL algorithms applied to EFGs as they stumble upon the inherent nonconvexity of the optimization landscape and the failure of the value-iteration subroutine in EFGs. To this goal, we utilize contemporary advances in nonconvex optimization theory to prove that regularized alternating policy gradient with (i) *direct policy parametrization*, (ii) *softmax policy parametrization*, and (iii) *softmax policy parametrization with natural policy gradient* updates converge to an approximate Nash equilibrium (NE) in the *last-iterate* in imperfect-information perfect-recall zero-sum EFGs. Namely, we observe that since the individual utilities are concave with respect to the sequence-form strategy, they satisfy gradient dominance w.r.t. the behavioral strategy---or, \textit{policy}, in reinforcement learning terms. We exploit this structure to further prove that the regularized utility satisfies the much stronger proximal Polyak- Łojasiewicz condition. In turn, we show that the different flavors of alternating policy gradient methods converge to an $\epsilon$-approximate NE with a number of iterations and trajectory samples that are polynomial in $1/\epsilon$ and the natural parameters of the game. Our work is a preliminary---yet principled---attempt in bridging the conceptual gap between the theory of Markov and imperfect-information EFGs while it aspires to stimulate a deeper dialogue between them.

Deep Learning · Graph Neural Networks

Xinya Qin, Lu Bai, Lixin Cui, Ming Li, Hangyuan Du, Edwin Hancock

The problem of over-smoothing has emerged as a fundamental issue for Graph Convolutional Networks (GCNs). While existing efforts primarily focus on enhancing the discriminability of node representations for node classification, they tend to overlook the over-smoothing at the graph level, significantly influencing the performance of graph classification. In this paper, we provide an explanation of the graph-level over-smoothing phenomenon, and propose a novel Adaptive Multi-Viewed Subgraph Convolutional Network (MultiNet) to address this challenge. Specifically, the MultiNet introduces a local subgraph convolution module that adaptively divides each input graph into multiple subgraph views. Then a number of subgraph-based view-specific convolution operations are applied to constrain the extent of node information propagation over the original global graph structure, not only mitigating the over-smoothing issue but also generating more discriminative local node representations. Moreover, we develop an alignment-based readout that establishes correspondences between nodes over different graphs, thereby effectively preserving the local node-level structure information and improving the discriminative ability of the resulting graph-level representations. Theoretical analysis and empirical studies show that the MultiNet mitigates the graph-level over-smoothing and achieves excellent performance for graph classification.

Reinforcement Learning · Multi-agent

Ruaridh Mon-Williams, Max Taylor-Davies, Elizabeth Mieczkowski, Natalia Vélez, Neil Bramley, Yanwei Wang, Tom Griffiths, Christopher G Lucas

Humans are remarkably adept at collaboration, able to infer the strengths and weaknesses of new partners in order to work successfully towards shared goals. To build AI systems with this capability, we must first understand its building blocks: does such flexibility require explicit, dedicated mechanisms for modelling others—or can it emerge spontaneously from the pressures of open-ended cooperative interaction? To investigate this question, we train simple model-free RNN agents to collaborate with a population of diverse partners. Using the 'Overcooked-AI' environment, we collect data from thousands of collaborative teams, and analyse agents' internal hidden states. Despite a lack of additional architectural features, inductive biases, or auxiliary objectives, the agents nevertheless develop structured internal representations of their partners' task abilities, enabling rapid adaptation and generalisation to novel collaborators. We investigated these internal models through probing techniques, and large-scale behavioural analysis. Notably, we find that structured partner modelling emerges when agents can influence partner behaviour by controlling task allocation. Our results show that partner modelling can arise spontaneously in model-free agents—but only under environmental conditions that impose the right kind of social pressure.

Reinforcement Learning · Deep RL

Xiaoliang Hu, Fuyun Wang, Tong Zhang, Zhen Cui

Model-free reinforcement learning (RL) combined with diffusion models has achieved significant progress in addressing complex continuous control tasks. However, a persistent challenge in RL remains the accurate estimation of Q-values, which critically governs the efficacy of policy optimization. Although recent advances employ parametric distributions to model value distributions for enhanced estimation accuracy, current methodologies predominantly rely on unimodal Gaussian assumptions or quantile representations. These constraints introduce distributional bias between the learned and true value distributions, particularly in some tasks with a nonstationary policy, ultimately degrading performance. To address these limitations, we propose value diffusion reinforcement learning (VDRL), a novel model-free online RL method that utilizes the generative capacity of diffusion models to represent multimodal value distributions. The core innovation of VDRL lies in the use of the variational loss of diffusion-based value distribution, which is theoretically proven to be a tight lower bound for the optimization objective under the KL-divergence measurement. Furthermore, we introduce double value diffusion learning with sample selection to enhance training stability and further improve value estimation accuracy. Extensive experiments conducted on the MuJoCo benchmark demonstrate that VDRL significantly outperforms some SOTA model-free online RL baselines, showcasing its effectiveness and robustness.

General Machine Learning · Representation Learning

Chong Tang, Joseph Powell, Dirk Koch, Robert Mullins, Alex Weddell, Jagmohan Chauhan

Recent progress on Remote Sensing Foundation Models (RSFMs) aims toward universal representations for Earth observation imagery. However, current efforts often scale up in size significantly without addressing efficiency constraints critical for real-world applications (e.g., onboard processing, rapid disaster response) or treat multispectral (MS) data as generic imagery, overlooking valuable physical priors. We introduce PhySwin, a foundation model for MS data that integrates physical priors with computational efficiency. PhySwin combines three innovations: (i) physics-informed pretraining objectives leveraging radiometric constraints to enhance feature learning; (ii) an efficient MixMAE formulation tailored to SwinV2 for low-FLOP, scalable pretraining; and (iii) token-efficient spectral embedding to retain spectral detail without increasing token counts. Pretrained on over 1M Sentinel-2 tiles, PhySwin achieves SOTA results (+1.32\% mIoU segmentation, +0.80\% F1 change detection) while reducing inference latency by up to 14.4$\times$ and computational complexity by up to 43.6$\times$ compared to ViT-based RSFMs.

Computer Vision · Image and Video Generation

Peiyuan Zhang, Yongqi Chen, Haofeng Huang, Will Lin, Zhengzhong Liu, Ion Stoica, Eric Xing, Hao Zhang

Scaling video diffusion transformers (DiTs) is limited by their quadratic 3D attention, even though most of the attention mass concentrates on a small subset of positions. We turn this observation into VSA, a trainable, hardware-efficient sparse attention that replaces full attention at both training and inference. In VSA, a lightweight coarse stage pools tokens into tiles and identifies high-weight critical tokens; a fine stage computes token-level attention only inside those tiles subjecting to block computing layout to ensure hard efficiency. This leads to a single differentiable kernel that trains end-to-end, requires no post-hoc profiling, and sustains 85\% of FlashAttention3 MFU. We perform a large sweep of ablation studies and scaling-law experiments by pretraining DiTs from 60M to 1.4B parameters. VSA reaches a Pareto point that cuts training FLOPS by 2.53$\times$ with no drop in diffusion loss. Retrofitting the open-source Wan2.1-1.3B model speeds up attention time by 6$\times$ and lowers end-to-end generation time from 31s to 18s with comparable quality, while for the 14B model, end-to-end generation time is reduced from 1274s to 576s. Furthermore, we introduce a preliminary study of Sparse-Distill, the first method to enable sparse attention and distillation concurrently, achieving 50.9x speed up for Wan-1.3B while maintaining quality. These results establish trainable sparse attention as a practical alternative to full attention and a key enabler for further scaling of video diffusion models. Code is available at https://github.com/hao-ai-lab/FastVideo.

General Machine Learning · Representation Learning

Ye Liu, Zihan Ji, Hongmin Cai

Multimodal representation learning is critical for a wide range of applications, such as multimodal sentiment analysis. Current multimodal representation learning methods mainly focus on the multimodal alignment or fusion strategies, such that the complementary and consistent information among heterogeneous modalities can be fully explored. However, they mistakenly treat the uncertainty noise within each modality as the complementary information, failing to simultaneously leverage both consistent and complementary information while eliminating the aleatoric uncertainty within each modality. To address this issue, we propose a plug-and-play feature causality decomposition method for multimodal representation learning from causality perspective, which can be integrated into existing models with no affects on the original model structures. Specifically, to deal with the heterogeneity and consistency, according to whether it can be aligned with other modalities, the unimodal feature is first disentangled into two parts: modality-invariant (the synergistic information shared by all heterogeneous modalities) and modality-specific part. To deal with complementarity and uncertainty, the modality-specific part is further decomposed into unique and redundant features, where the redundant feature is removed and the unique feature is reserved based on the backdoor-adjustment. The effectiveness of noise removal is supported by causality theory. Finally, the task-related information, including both synergistic and unique components, is further fed to the original fusion module to obtain the final multimodal representations. Extensive experiments show the effectiveness of our proposed strategies.

Applications · Language, Speech and Dialog

Zeqian Ju, Dongchao Yang, Shen Kai, YICHONG LENG, Zhengtao Wang, Songxiang Liu, Xinyu Zhou, Tao Qin, Xiangyang Li, Jianwei Yu 等

Recent advances in text-to-speech synthesis have achieved notable success in generating high-quality short utterances for individual speakers. However, these systems still face challenges when extending their capabilities to long, multi-speaker, and spontaneous dialogues, typical of real-world scenarios such as podcasts. These limitations arise from two primary challenges: 1) long speech: podcasts typically span several minutes, exceeding the upper limit of most existing work; 2) spontaneity: podcasts are marked by their spontaneous, oral nature, which sharply contrasts with formal, written contexts; existing works often fall short in capturing this spontaneity. In this paper, we propose MoonCast, a solution for high-quality zero-shot podcast generation, aiming to synthesize spontaneous podcast-style speech from text-only sources (e.g., stories, technical reports, news in TXT, PDF, or Web URL formats) using the voices of unseen speakers. To enable long audio generation, we employ a language model with parameter, data, and context scaling to process sequences in an innovative format designed for modeling entire multi-speaker, multi-turn speech interactions. To enhance spontaneity, we observe that ASR transcripts capture spontaneous speech details (e.g., filler words indicating hesitations, and specific punctuation and spaces reflecting breathing pauses), suggesting that these transcripts can serve as a partial indicator of speech spontaneity. Building upon this assumption, we utilize a script generation module to generate scripts incorporating these spontaneous elements. Experiments show MoonCast outperforms baselines, with notable improvements in contextual coherence and spontaneity.