Zhipeng Hu

Yifeng Ma, Suzhen Wang, Zhipeng Hu et al.

Different people speak with diverse personalized speaking styles. Although existing one-shot talking head methods have made significant progress in lip sync, natural facial expressions, and stable head motions, they still cannot generate diverse speaking styles in the final talking head videos. To tackle this problem, we propose a one-shot style-controllable talking face generation framework. In a nutshell, we aim to attain a speaking style from an arbitrary reference speaking video and then drive the one-shot portrait to speak with the reference speaking style and another piece of audio. Specifically, we first develop a style encoder to extract dynamic facial motion patterns of a style reference video and then encode them into a style code. Afterward, we introduce a style-controllable decoder to synthesize stylized facial animations from the speech content and style code. In order to integrate the reference speaking style into generated videos, we design a style-aware adaptive transformer, which enables the encoded style code to adjust the weights of the feed-forward layers accordingly. Thanks to the style-aware adaptation mechanism, the reference speaking style can be better embedded into synthesized videos during decoding. Extensive experiments demonstrate that our method is capable of generating talking head videos with diverse speaking styles from only one portrait image and an audio clip while achieving authentic visual effects. Project Page: https://github.com/FuxiVirtualHuman/styletalk.

Haozhe Ji, Pei Ke, Zhipeng Hu et al. · tsinghua

The standard paradigm of neural language generation adopts maximum likelihood estimation (MLE) as the optimizing method. From a distributional view, MLE in fact minimizes the Kullback-Leibler divergence (KLD) between the distribution of the real data and that of the model. However, this approach forces the model to distribute non-zero (sometimes large) probability mass to all training samples regardless of their quality. Moreover, in the attempt to cover the low-probability regions in the data distribution, the model systematically overestimates the probability of corrupted text sequences, which we conjecture is one of the main reasons for text degeneration during autoregressive decoding. To remedy this problem, we leverage the total variation distance (TVD) with its robustness to outliers, and develop practical bounds to apply it to language generation. Then, we introduce the TaiLr objective that balances the tradeoff of estimating TVD. Intuitively, TaiLr downweights real data samples that have low model probabilities with tunable penalization intensity. Experimental results show that our method alleviates the overestimation of degenerated sequences without sacrificing diversity and improves generation quality on a wide range of text generation tasks.

Jian Guan, Zhenyu Yang, Rongsheng Zhang et al. · tsinghua

Despite advances in generating fluent texts, existing pretraining models tend to attach incoherent event sequences to involved entities when generating narratives such as stories and news. We conjecture that such issues result from representing entities as static embeddings of superficial words, while neglecting to model their ever-changing states, i.e., the information they carry, as the text unfolds. Therefore, we extend the Transformer model to dynamically conduct entity state updates and sentence realization for narrative generation. We propose a contrastive framework to learn the state representations in a discrete space, and insert additional attention layers into the decoder to better exploit these states. Experiments on two narrative datasets show that our model can generate more coherent and diverse narratives than strong baselines with the guidance of meaningful entity states.

Haozhe Ji, Rongsheng Zhang, Zhenyu Yang et al. · tsinghua

Although Transformers with fully connected self-attentions are powerful to model long-term dependencies, they are struggling to scale to long texts with thousands of words in language modeling. One of the solutions is to equip the model with a recurrence memory. However, existing approaches directly reuse hidden states from the previous segment that encodes contexts in a uni-directional way. As a result, this prohibits the memory to dynamically interact with the current context that provides up-to-date information for token prediction. To remedy this issue, we propose Look-Ahead Memory (LaMemo) that enhances the recurrence memory by incrementally attending to the right-side tokens, and interpolating with the old memory states to maintain long-term information in the history. LaMemo embraces bi-directional attention and segment recurrence with an additional computation overhead only linearly proportional to the memory length. Experiments on widely used language modeling benchmarks demonstrate its superiority over the baselines equipped with different types of memory.

Zhimeng Zhang, Zhipeng Hu, Wenjin Deng et al.

For few-shot learning, it is still a critical challenge to realize photo-realistic face visually dubbing on high-resolution videos. Previous works fail to generate high-fidelity dubbing results. To address the above problem, this paper proposes a Deformation Inpainting Network (DINet) for high-resolution face visually dubbing. Different from previous works relying on multiple up-sample layers to directly generate pixels from latent embeddings, DINet performs spatial deformation on feature maps of reference images to better preserve high-frequency textural details. Specifically, DINet consists of one deformation part and one inpainting part. In the first part, five reference facial images adaptively perform spatial deformation to create deformed feature maps encoding mouth shapes at each frame, in order to align with the input driving audio and also the head poses of the input source images. In the second part, to produce face visually dubbing, a feature decoder is responsible for adaptively incorporating mouth movements from the deformed feature maps and other attributes (i.e., head pose and upper facial expression) from the source feature maps together. Finally, DINet achieves face visually dubbing with rich textural details. We conduct qualitative and quantitative comparisons to validate our DINet on high-resolution videos. The experimental results show that our method outperforms state-of-the-art works.

Haoqian Wu, Zhipeng Hu, Lincheng Li et al.

Inverse rendering methods aim to estimate geometry, materials and illumination from multi-view RGB images. In order to achieve better decomposition, recent approaches attempt to model indirect illuminations reflected from different materials via Spherical Gaussians (SG), which, however, tends to blur the high-frequency reflection details. In this paper, we propose an end-to-end inverse rendering pipeline that decomposes materials and illumination from multi-view images, while considering near-field indirect illumination. In a nutshell, we introduce the Monte Carlo sampling based path tracing and cache the indirect illumination as neural radiance, enabling a physics-faithful and easy-to-optimize inverse rendering method. To enhance efficiency and practicality, we leverage SG to represent the smooth environment illuminations and apply importance sampling techniques. To supervise indirect illuminations from unobserved directions, we develop a novel radiance consistency constraint between implicit neural radiance and path tracing results of unobserved rays along with the joint optimization of materials and illuminations, thus significantly improving the decomposition performance. Extensive experiments demonstrate that our method outperforms the state-of-the-art on multiple synthetic and real datasets, especially in terms of inter-reflection decomposition.Our code and data are available at https://woolseyyy.github.io/nefii/.

Rui Zhao, Wei Li, Zhipeng Hu et al.

Recent popular Role-Playing Games (RPGs) saw the great success of character auto-creation systems. The bone-driven face model controlled by continuous parameters (like the position of bones) and discrete parameters (like the hairstyles) makes it possible for users to personalize and customize in-game characters. Previous in-game character auto-creation systems are mostly image-driven, where facial parameters are optimized so that the rendered character looks similar to the reference face photo. This paper proposes a novel text-to-parameter translation method (T2P) to achieve zero-shot text-driven game character auto-creation. With our method, users can create a vivid in-game character with arbitrary text description without using any reference photo or editing hundreds of parameters manually. In our method, taking the power of large-scale pre-trained multi-modal CLIP and neural rendering, T2P searches both continuous facial parameters and discrete facial parameters in a unified framework. Due to the discontinuous parameter representation, previous methods have difficulty in effectively learning discrete facial parameters. T2P, to our best knowledge, is the first method that can handle the optimization of both discrete and continuous parameters. Experimental results show that T2P can generate high-quality and vivid game characters with given text prompts. T2P outperforms other SOTA text-to-3D generation methods on both objective evaluations and subjective evaluations.

Yifeng Ma, Suzhen Wang, Yu Ding et al.

Audio-driven talking head generation has drawn growing attention. To produce talking head videos with desired facial expressions, previous methods rely on extra reference videos to provide expression information, which may be difficult to find and hence limits their usage. In this work, we propose TalkCLIP, a framework that can generate talking heads where the expressions are specified by natural language, hence allowing for specifying expressions more conveniently. To model the mapping from text to expressions, we first construct a text-video paired talking head dataset where each video has diverse text descriptions that depict both coarse-grained emotions and fine-grained facial movements. Leveraging the proposed dataset, we introduce a CLIP-based style encoder that projects natural language-based descriptions to the representations of expressions. TalkCLIP can even infer expressions for descriptions unseen during training. TalkCLIP can also use text to modulate expression intensity and edit expressions. Extensive experiments demonstrate that TalkCLIP achieves the advanced capability of generating photo-realistic talking heads with vivid facial expressions guided by text descriptions.

Bowen Ma, Rudong An, Wei Zhang et al.

As a fine-grained and local expression behavior measurement, facial action unit (FAU) analysis (e.g., detection and intensity estimation) has been documented for its time-consuming, labor-intensive, and error-prone annotation. Thus a long-standing challenge of FAU analysis arises from the data scarcity of manual annotations, limiting the generalization ability of trained models to a large extent. Amounts of previous works have made efforts to alleviate this issue via semi/weakly supervised methods and extra auxiliary information. However, these methods still require domain knowledge and have not yet avoided the high dependency on data annotation. This paper introduces a robust facial representation model MAE-Face for AU analysis. Using masked autoencoding as the self-supervised pre-training approach, MAE-Face first learns a high-capacity model from a feasible collection of face images without additional data annotations. Then after being fine-tuned on AU datasets, MAE-Face exhibits convincing performance for both AU detection and AU intensity estimation, achieving a new state-of-the-art on nearly all the evaluation results. Further investigation shows that MAE-Face achieves decent performance even when fine-tuned on only 1\% of the AU training set, strongly proving its robustness and generalization performance.

Zibin Dong, Yifu Yuan, Jianye Hao et al.

Aligning agent behaviors with diverse human preferences remains a challenging problem in reinforcement learning (RL), owing to the inherent abstractness and mutability of human preferences. To address these issues, we propose AlignDiff, a novel framework that leverages RL from Human Feedback (RLHF) to quantify human preferences, covering abstractness, and utilizes them to guide diffusion planning for zero-shot behavior customizing, covering mutability. AlignDiff can accurately match user-customized behaviors and efficiently switch from one to another. To build the framework, we first establish the multi-perspective human feedback datasets, which contain comparisons for the attributes of diverse behaviors, and then train an attribute strength model to predict quantified relative strengths. After relabeling behavioral datasets with relative strengths, we proceed to train an attribute-conditioned diffusion model, which serves as a planner with the attribute strength model as a director for preference aligning at the inference phase. We evaluate AlignDiff on various locomotion tasks and demonstrate its superior performance on preference matching, switching, and covering compared to other baselines. Its capability of completing unseen downstream tasks under human instructions also showcases the promising potential for human-AI collaboration. More visualization videos are released on https://aligndiff.github.io/.

Suzhen Wang, Yifeng Ma, Yu Ding et al.

Individuals have unique facial expression and head pose styles that reflect their personalized speaking styles. Existing one-shot talking head methods cannot capture such personalized characteristics and therefore fail to produce diverse speaking styles in the final videos. To address this challenge, we propose a one-shot style-controllable talking face generation method that can obtain speaking styles from reference speaking videos and drive the one-shot portrait to speak with the reference speaking styles and another piece of audio. Our method aims to synthesize the style-controllable coefficients of a 3D Morphable Model (3DMM), including facial expressions and head movements, in a unified framework. Specifically, the proposed framework first leverages a style encoder to extract the desired speaking styles from the reference videos and transform them into style codes. Then, the framework uses a style-aware decoder to synthesize the coefficients of 3DMM from the audio input and style codes. During decoding, our framework adopts a two-branch architecture, which generates the stylized facial expression coefficients and stylized head movement coefficients, respectively. After obtaining the coefficients of 3DMM, an image renderer renders the expression coefficients into a specific person's talking-head video. Extensive experiments demonstrate that our method generates visually authentic talking head videos with diverse speaking styles from only one portrait image and an audio clip.

Zhipeng Hu, Wei Zhang, Lincheng Li et al.

Since Facial Action Unit (AU) annotations require domain expertise, common AU datasets only contain a limited number of subjects. As a result, a crucial challenge for AU detection is addressing identity overfitting. We find that AUs and facial expressions are highly associated, and existing facial expression datasets often contain a large number of identities. In this paper, we aim to utilize the expression datasets without AU labels to facilitate AU detection. Specifically, we develop a novel AU detection framework aided by the Global-Local facial Expressions Embedding, dubbed GLEE-Net. Our GLEE-Net consists of three branches to extract identity-independent expression features for AU detection. We introduce a global branch for modeling the overall facial expression while eliminating the impacts of identities. We also design a local branch focusing on specific local face regions. The combined output of global and local branches is firstly pre-trained on an expression dataset as an identity-independent expression embedding, and then finetuned on AU datasets. Therefore, we significantly alleviate the issue of limited identities. Furthermore, we introduce a 3D global branch that extracts expression coefficients through 3D face reconstruction to consolidate 2D AU descriptions. Finally, a Transformer-based multi-label classifier is employed to fuse all the representations for AU detection. Extensive experiments demonstrate that our method significantly outperforms the state-of-the-art on the widely-used DISFA, BP4D and BP4D+ datasets.

Pengfei Xi, Guifeng Wang, Zhipeng Hu et al.

Determining causal effects of temporal multi-intervention assists decision-making. Restricted by time-varying bias, selection bias, and interactions of multiple interventions, the disentanglement and estimation of multiple treatment effects from individual temporal data is still rare. To tackle these challenges, we propose a comprehensive framework of temporal counterfactual forecasting from an individual multiple treatment perspective (TCFimt). TCFimt constructs adversarial tasks in a seq2seq framework to alleviate selection and time-varying bias and designs a contrastive learning-based block to decouple a mixed treatment effect into separated main treatment effects and causal interactions which further improves estimation accuracy. Through implementing experiments on two real-world datasets from distinct fields, the proposed method shows satisfactory performance in predicting future outcomes with specific treatments and in choosing optimal treatment type and timing than state-of-the-art methods.

Feng Qiu, Wei Zhang, Chen Liu et al.

Video-driven 3D facial animation transfer aims to drive avatars to reproduce the expressions of actors. Existing methods have achieved remarkable results by constraining both geometric and perceptual consistency. However, geometric constraints (like those designed on facial landmarks) are insufficient to capture subtle emotions, while expression features trained on classification tasks lack fine granularity for complex emotions. To address this, we propose \textbf{FreeAvatar}, a robust facial animation transfer method that relies solely on our learned expression representation. Specifically, FreeAvatar consists of two main components: the expression foundation model and the facial animation transfer model. In the first component, we initially construct a facial feature space through a face reconstruction task and then optimize the expression feature space by exploring the similarities among different expressions. Benefiting from training on the amounts of unlabeled facial images and re-collected expression comparison dataset, our model adapts freely and effectively to any in-the-wild input facial images. In the facial animation transfer component, we propose a novel Expression-driven Multi-avatar Animator, which first maps expressive semantics to the facial control parameters of 3D avatars and then imposes perceptual constraints between the input and output images to maintain expression consistency. To make the entire process differentiable, we employ a trained neural renderer to translate rig parameters into corresponding images. Furthermore, unlike previous methods that require separate decoders for each avatar, we propose a dynamic identity injection module that allows for the joint training of multiple avatars within a single network.

Zhipeng Hu, Minda Zhao, Chaoyi Zhao et al.

While image diffusion models have made significant progress in text-driven 3D content creation, they often fail to accurately capture the intended meaning of text prompts, especially for view information. This limitation leads to the Janus problem, where multi-faced 3D models are generated under the guidance of such diffusion models. In this paper, we propose a robust high-quality 3D content generation pipeline by exploiting orthogonal-view image guidance. First, we introduce a novel 2D diffusion model that generates an image consisting of four orthogonal-view sub-images based on the given text prompt. Then, the 3D content is created using this diffusion model. Notably, the generated orthogonal-view image provides strong geometric structure priors and thus improves 3D consistency. As a result, it effectively resolves the Janus problem and significantly enhances the quality of 3D content creation. Additionally, we present a 3D synthesis fusion network that can further improve the details of the generated 3D contents. Both quantitative and qualitative evaluations demonstrate that our method surpasses previous text-to-3D techniques. Project page: https://efficientdreamer.github.io.

Yuhang Ma, Wenting Xu, Chaoyi Zhao et al.

Recent advances in text-to-image diffusion models have spurred significant interest in continuous story image generation. In this paper, we introduce Storynizor, a model capable of generating coherent stories with strong inter-frame character consistency, effective foreground-background separation, and diverse pose variation. The core innovation of Storynizor lies in its key modules: ID-Synchronizer and ID-Injector. The ID-Synchronizer employs an auto-mask self-attention module and a mask perceptual loss across inter-frame images to improve the consistency of character generation, vividly representing their postures and backgrounds. The ID-Injector utilize a Shuffling Reference Strategy (SRS) to integrate ID features into specific locations, enhancing ID-based consistent character generation. Additionally, to facilitate the training of Storynizor, we have curated a novel dataset called StoryDB comprising 100, 000 images. This dataset contains single and multiple-character sets in diverse environments, layouts, and gestures with detailed descriptions. Experimental results indicate that Storynizor demonstrates superior coherent story generation with high-fidelity character consistency, flexible postures, and vivid backgrounds compared to other character-specific methods.

Jiawei Wang, Kai Wang, Shaojie Lin et al.

With the rapid advancement of Large Language Models (LLMs), LLM-based autonomous agents have shown the potential to function as digital employees, such as digital analysts, teachers, and programmers. In this paper, we develop an application-level testbed based on the open-source strategy game "Unciv", which has millions of active players, to enable researchers to build a "data flywheel" for studying human-like agents in the "digital players" task. This "Civilization"-like game features expansive decision-making spaces along with rich linguistic interactions such as diplomatic negotiations and acts of deception, posing significant challenges for LLM-based agents in terms of numerical reasoning and long-term planning. Another challenge for "digital players" is to generate human-like responses for social interaction, collaboration, and negotiation with human players. The open-source project can be found at https:/github.com/fuxiAIlab/CivAgent.

Yu Xiong, Zhipeng Hu, Ye Huang et al.

Reinforcement Learning (RL) has demonstrated substantial potential across diverse fields, yet understanding its decision-making process, especially in real-world scenarios where rationality and safety are paramount, is an ongoing challenge. This paper delves in to Explainable RL (XRL), a subfield of Explainable AI (XAI) aimed at unravelling the complexities of RL models. Our focus rests on state-explaining techniques, a crucial subset within XRL methods, as they reveal the underlying factors influencing an agent's actions at any given time. Despite their significant role, the lack of a unified evaluation framework hinders assessment of their accuracy and effectiveness. To address this, we introduce XRL-Bench, a unified standardized benchmark tailored for the evaluation and comparison of XRL methods, encompassing three main modules: standard RL environments, explainers based on state importance, and standard evaluators. XRL-Bench supports both tabular and image data for state explanation. We also propose TabularSHAP, an innovative and competitive XRL method. We demonstrate the practical utility of TabularSHAP in real-world online gaming services and offer an open-source benchmark platform for the straightforward implementation and evaluation of XRL methods. Our contributions facilitate the continued progression of XRL technology.

Wei Chen, Liangmin Wu, Yunhai Hu et al.

While Neural Processing Units (NPUs) offer high theoretical efficiency for edge AI, state-of-the-art Vision--Language Models (VLMs) tailored for GPUs often falter on these substrates. We attribute this hardware-model mismatch to two primary factors: the quantization brittleness of Vision Transformers (ViTs) and the I/O-bound nature of autoregressive attention mechanisms, which fail to utilize the high arithmetic throughput of NPUs. To bridge this gap, we propose AutoNeural, an NPU-native VLM architecture co-designed for integer-only inference. We replace the standard ViT encoder with a MobileNetV5-style backbone utilizing depthwise separable convolutions, which ensures bounded activation distributions for stable INT4/8/16 quantization. Complementing this, our language backbone integrates State-Space Model (SSM) principles with Transformer layers, employing efficient gated convolutions to achieve linear-time complexity. This hybrid design eliminates the heavy memory I/O overhead of Key-Value caching during generation. Our approach delivers substantial efficiency gains, reducing quantization error of vision encoder by up to 7x and end-to-end latency by 14x compared to conventional baselines. The AutoNeural also delivers 3x decoding speed and 4x longer context window than the baseline. We validate these improvements via a real-world automotive case study on the Qualcomm SA8295P SoC, demonstrating real-time performance for cockpit applications. Our results highlight that rethinking model topology specifically for NPU constraints is a prerequisite for robust multi-modal edge intelligence.

Oucheng Huang, Yuhang Ma, Zeng Zhao et al.

ComfyUI is a popular workflow-based interface that allows users to customize image generation tasks through an intuitive node-based system. However, the complexity of managing node connections and diverse modules can be challenging for users. In this paper, we introduce ComfyGPT, a self-optimizing multi-agent system designed to generate ComfyUI workflows based on task descriptions automatically. The key innovations of ComfyGPT include: (1) consisting of four specialized agents to build a multi-agent workflow generation system: ReformatAgent, FlowAgent, RefineAgent, and ExecuteAgent; (2) focusing on generating precise node connections instead of entire workflows, improving generation accuracy; and (3) enhancing workflow generation through reinforcement learning. Moreover, we introduce FlowDataset, a large-scale dataset containing 13,571 workflow-description pairs, and FlowBench, a comprehensive benchmark for evaluating workflow generation systems. Additionally, we propose four novel evaluation metrics: Format Validation (FV), Pass Accuracy (PA), Pass Instruct Alignment (PIA), and Pass Node Diversity (PND). Experimental results demonstrate that ComfyGPT significantly outperforms existing LLM-based methods in workflow generation, making it a significant step forward in this field. Code is avaliable at https://github.com/comfygpt/comfygpt.

Xiaoyan Kui, Canwei Liu, Qinsong Li et al.

Kolmogorov-Arnold Networks (KANs) are highly effective in long-term time series forecasting due to their ability to efficiently represent nonlinear relationships and exhibit local plasticity. However, prior research on KANs has predominantly focused on the time domain, neglecting the potential of the frequency domain. The frequency domain of time series data reveals recurring patterns and periodic behaviors, which complement the temporal information captured in the time domain. To address this gap, we explore the application of KANs in the frequency domain for long-term time series forecasting. By leveraging KANs' adaptive activation functions and their comprehensive representation of signals in the frequency domain, we can more effectively learn global dependencies and periodic patterns. To integrate information from both time and frequency domains, we propose the $\textbf{T}$ime-$\textbf{F}$requency KAN (TFKAN). TFKAN employs a dual-branch architecture that independently processes features from each domain, ensuring that the distinct characteristics of each domain are fully utilized without interference. Additionally, to account for the heterogeneity between domains, we introduce a dimension-adjustment strategy that selectively upscales only in the frequency domain, enhancing efficiency while capturing richer frequency information. Experimental results demonstrate that TFKAN consistently outperforms state-of-the-art (SOTA) methods across multiple datasets. The code is available at https://github.com/LcWave/TFKAN.

Yuhang Ma, Wenting Xu, Jiji Tang et al.

Customized image generation, which seeks to synthesize images with consistent characters, holds significant relevance for applications such as storytelling, portrait generation, and character design. However, previous approaches have encountered challenges in preserving characters with high-fidelity consistency due to inadequate feature extraction and concept confusion of reference characters. Therefore, we propose Character-Adapter, a plug-and-play framework designed to generate images that preserve the details of reference characters, ensuring high-fidelity consistency. Character-Adapter employs prompt-guided segmentation to ensure fine-grained regional features of reference characters and dynamic region-level adapters to mitigate concept confusion. Extensive experiments are conducted to validate the effectiveness of Character-Adapter. Both quantitative and qualitative results demonstrate that Character-Adapter achieves the state-of-the-art performance of consistent character generation, with an improvement of 24.8% compared with other methods. Our code will be released at https://github.com/Character-Adapter/Character-Adapter.

Jinfeng Zhou, Yongkang Huang, Bosi Wen et al.

Character-based dialogue (aka role-playing) enables users to freely customize characters for interaction, which often relies on LLMs, raising the need to evaluate LLMs' character customization capability. However, existing benchmarks fail to ensure a robust evaluation as they often only involve a single character category or evaluate limited dimensions. Moreover, the sparsity of character features in responses makes feature-focused generative evaluation both ineffective and inefficient. To address these issues, we propose CharacterBench, the largest bilingual generative benchmark, with 22,859 human-annotated samples covering 3,956 characters from 25 detailed character categories. We define 11 dimensions of 6 aspects, classified as sparse and dense dimensions based on whether character features evaluated by specific dimensions manifest in each response. We enable effective and efficient evaluation by crafting tailored queries for each dimension to induce characters' responses related to specific dimensions. Further, we develop CharacterJudge model for cost-effective and stable evaluations. Experiments show its superiority over SOTA automatic judges (e.g., GPT-4) and our benchmark's potential to optimize LLMs' character customization. Our repository is at https://github.com/thu-coai/CharacterBench.

Yufeng Huang, Jiji Tang, Zhuo Chen et al.

Large-scale vision-language pre-training has achieved significant performance in multi-modal understanding and generation tasks. However, existing methods often perform poorly on image-text matching tasks that require structured representations, i.e., representations of objects, attributes, and relations. As illustrated in Fig.~reffig:case (a), the models cannot make a distinction between ``An astronaut rides a horse" and ``A horse rides an astronaut". This is because they fail to fully leverage structured knowledge when learning representations in multi-modal scenarios. In this paper, we present an end-to-end framework Structure-CLIP, which integrates Scene Graph Knowledge (SGK) to enhance multi-modal structured representations. Firstly, we use scene graphs to guide the construction of semantic negative examples, which results in an increased emphasis on learning structured representations. Moreover, a Knowledge-Enhance Encoder (KEE) is proposed to leverage SGK as input to further enhance structured representations. To verify the effectiveness of the proposed framework, we pre-train our model with the aforementioned approaches and conduct experiments on downstream tasks. Experimental results demonstrate that Structure-CLIP achieves state-of-the-art (SOTA) performance on VG-Attribution and VG-Relation datasets, with 12.5% and 4.1% ahead of the multi-modal SOTA model respectively. Meanwhile, the results on MSCOCO indicate that Structure-CLIP significantly enhances the structured representations while maintaining the ability of general representations. Our code is available at https://github.com/zjukg/Structure-CLIP.

Renshuai Liu, Bowen Ma, Wei Zhang et al.

In human-centric content generation, the pre-trained text-to-image models struggle to produce user-wanted portrait images, which retain the identity of individuals while exhibiting diverse expressions. This paper introduces our efforts towards personalized face generation. To this end, we propose a novel multi-modal face generation framework, capable of simultaneous identity-expression control and more fine-grained expression synthesis. Our expression control is so sophisticated that it can be specialized by the fine-grained emotional vocabulary. We devise a novel diffusion model that can undertake the task of simultaneously face swapping and reenactment. Due to the entanglement of identity and expression, it's nontrivial to separately and precisely control them in one framework, thus has not been explored yet. To overcome this, we propose several innovative designs in the conditional diffusion model, including balancing identity and expression encoder, improved midpoint sampling, and explicitly background conditioning. Extensive experiments have demonstrated the controllability and scalability of the proposed framework, in comparison with state-of-the-art text-to-image, face swapping, and face reenactment methods.

Siyu Zou, Jiji Tang, Yiyi Zhou et al.

Diffusion-based Image Editing (DIE) is an emerging research hot-spot, which often applies a semantic mask to control the target area for diffusion-based editing. However, most existing solutions obtain these masks via manual operations or off-line processing, greatly reducing their efficiency. In this paper, we propose a novel and efficient image editing method for Text-to-Image (T2I) diffusion models, termed Instant Diffusion Editing(InstDiffEdit). In particular, InstDiffEdit aims to employ the cross-modal attention ability of existing diffusion models to achieve instant mask guidance during the diffusion steps. To reduce the noise of attention maps and realize the full automatics, we equip InstDiffEdit with a training-free refinement scheme to adaptively aggregate the attention distributions for the automatic yet accurate mask generation. Meanwhile, to supplement the existing evaluations of DIE, we propose a new benchmark called Editing-Mask to examine the mask accuracy and local editing ability of existing methods. To validate InstDiffEdit, we also conduct extensive experiments on ImageNet and Imagen, and compare it with a bunch of the SOTA methods. The experimental results show that InstDiffEdit not only outperforms the SOTA methods in both image quality and editing results, but also has a much faster inference speed, i.e., +5 to +6 times.

Suzhen Wang, Weijie Chen, Wei Zhang et al.

Character customization, or 'face crafting,' is a vital feature in role-playing games (RPGs), enhancing player engagement by enabling the creation of personalized avatars. Existing automated methods often struggle with generalizability across diverse game engines due to their reliance on the intermediate constraints of specific image domain and typically support only one type of input, either text or image. To overcome these challenges, we introduce EasyCraft, an innovative end-to-end feedforward framework that automates character crafting by uniquely supporting both text and image inputs. Our approach employs a translator capable of converting facial images of any style into crafting parameters. We first establish a unified feature distribution in the translator's image encoder through self-supervised learning on a large-scale dataset, enabling photos of any style to be embedded into a unified feature representation. Subsequently, we map this unified feature distribution to crafting parameters specific to a game engine, a process that can be easily adapted to most game engines and thus enhances EasyCraft's generalizability. By integrating text-to-image techniques with our translator, EasyCraft also facilitates precise, text-based character crafting. EasyCraft's ability to integrate diverse inputs significantly enhances the versatility and accuracy of avatar creation. Extensive experiments on two RPG games demonstrate the effectiveness of our method, achieving state-of-the-art results and facilitating adaptability across various avatar engines.

Shijing Zhu, Zhuang Chen, Guanqun Bi et al.

Large language models (LLMs) have shown promise in providing scalable mental health support, while evaluating their counseling capability remains crucial to ensure both efficacy and safety. Existing evaluations are limited by the static assessment that focuses on knowledge tests, the single perspective that centers on user experience, and the open-loop framework that lacks actionable feedback. To address these issues, we propose Ψ-Arena, an interactive framework for comprehensive assessment and optimization of LLM-based counselors, featuring three key characteristics: (1) Realistic arena interactions that simulate real-world counseling through multi-stage dialogues with psychologically profiled NPC clients, (2) Tripartite evaluation that integrates assessments from the client, counselor, and supervisor perspectives, and (3) Closed-loop optimization that iteratively improves LLM counselors using diagnostic feedback. Experiments across eight state-of-the-art LLMs show significant performance variations in different real-world scenarios and evaluation perspectives. Moreover, reflection-based optimization results in up to a 141% improvement in counseling performance. We hope PsychoArena provides a foundational resource for advancing reliable and human-aligned LLM applications in mental healthcare.

Jinfeng Zhou, Yuxuan Chen, Jianing Yin et al.

Cognitive Restructuring (CR) is a psychotherapeutic process aimed at identifying and restructuring an individual's negative thoughts, arising from mental health challenges, into more helpful and positive ones via multi-turn dialogues. Clinician shortage and stigma urge the development of human-LLM interactive psychotherapy for CR. Yet, existing efforts implement CR via simple text rewriting, fixed-pattern dialogues, or a one-shot CR workflow, failing to align with the psychotherapeutic process for effective CR. To address this gap, we propose CRDial, a novel framework for CR, which creates multi-turn dialogues with specifically designed identification and restructuring stages of negative thoughts, integrates sentence-level supportive conversation strategies, and adopts a multi-channel loop mechanism to enable iterative CR. With CRDial, we distill Crisp, a large-scale and high-quality bilingual dialogue dataset, from LLM. We then train Crispers, Crisp-based conversational LLMs for CR, at 7B and 14B scales. Extensive human studies show the superiority of Crispers in pointwise, pairwise, and intervention evaluations.

Bihan Xu, Shiwei Zhao, Runze Wu et al.

Within the domain of Massively Multiplayer Online (MMO) economy research, Agent-Based Modeling (ABM) has emerged as a robust tool for analyzing game economics, evolving from rule-based agents to decision-making agents enhanced by reinforcement learning. Nevertheless, existing works encounter significant challenges when attempting to emulate human-like economic activities among agents, particularly regarding agent reliability, sociability, and interpretability. In this study, we take a preliminary step in introducing a novel approach using Large Language Models (LLMs) in MMO economy simulation. Leveraging LLMs' role-playing proficiency, generative capacity, and reasoning aptitude, we design LLM-driven agents with human-like decision-making and adaptability. These agents are equipped with the abilities of role-playing, perception, memory, and reasoning, addressing the aforementioned challenges effectively. Simulation experiments focusing on in-game economic activities demonstrate that LLM-empowered agents can promote emergent phenomena like role specialization and price fluctuations in line with market rules.

Mushui Liu, Yuhang Ma, Yang Zhen et al.

Diffusion models have exhibited substantial success in text-to-image generation. However, they often encounter challenges when dealing with complex and dense prompts involving multiple objects, attribute binding, and long descriptions. In this paper, we propose a novel framework called \textbf{LLM4GEN}, which enhances the semantic understanding of text-to-image diffusion models by leveraging the representation of Large Language Models (LLMs). It can be seamlessly incorporated into various diffusion models as a plug-and-play component. A specially designed Cross-Adapter Module (CAM) integrates the original text features of text-to-image models with LLM features, thereby enhancing text-to-image generation. Additionally, to facilitate and correct entity-attribute relationships in text prompts, we develop an entity-guided regularization loss to further improve generation performance. We also introduce DensePrompts, which contains $7,000$ dense prompts to provide a comprehensive evaluation for the text-to-image generation task. Experiments indicate that LLM4GEN significantly improves the semantic alignment of SD1.5 and SDXL, demonstrating increases of 9.69\% and 12.90\% in color on T2I-CompBench, respectively. Moreover, it surpasses existing models in terms of sample quality, image-text alignment, and human evaluation.

Ziyang Luo, Zhipeng Hu, Yadong Xi et al.

Image Captioning is a traditional vision-and-language task that aims to generate the language description of an image. Recent studies focus on scaling up the model size and the number of training data, which significantly increase the cost of model training. Different to these heavy-cost models, we introduce a lightweight image captioning framework (I-Tuning), which contains a small number of trainable parameters. We design a novel I-Tuning cross-attention module to connect the non-trainable pre-trained language decoder GPT2 and vision encoder CLIP-ViT. Since most parameters are not required to be updated during training, our framework is lightweight and fast. Experimental results conducted on three image captioning benchmarks reveal that our framework achieves comparable or better performance than the large-scale baseline systems. But our models contain up to 10 times fewer trainable parameters and require much fewer data for training compared with state-of-the-art baselines.

Zhao-Hua Li, Yang Yu, Yingfeng Chen et al.

While deep reinforcement learning has achieved promising results in challenging decision-making tasks, the main bones of its success --- deep neural networks are mostly black-boxes. A feasible way to gain insight into a black-box model is to distill it into an interpretable model such as a decision tree, which consists of if-then rules and is easy to grasp and be verified. However, the traditional model distillation is usually a supervised learning task under a stationary data distribution assumption, which is violated in reinforcement learning. Therefore, a typical policy distillation that clones model behaviors with even a small error could bring a data distribution shift, resulting in an unsatisfied distilled policy model with low fidelity or low performance. In this paper, we propose to address this issue by changing the distillation objective from behavior cloning to maximizing an advantage evaluation. The novel distillation objective maximizes an approximated cumulative reward and focuses more on disastrous behaviors in critical states, which controls the data shift effect. We evaluate our method on several Gym tasks, a commercial fight game, and a self-driving car simulator. The empirical results show that the proposed method can preserve a higher cumulative reward than behavior cloning and learn a more consistent policy to the original one. Moreover, by examining the extracted rules from the distilled decision trees, we demonstrate that the proposed method delivers reasonable and robust decisions.

Ke Chen, Yufei Li, Yingfeng Chen et al.

Graphically-rich applications such as games are ubiquitous with attractive visual effects of Graphical User Interface (GUI) that offers a bridge between software applications and end-users. However, various types of graphical glitches may arise from such GUI complexity and have become one of the main component of software compatibility issues. Our study on bug reports from game development teams in NetEase Inc. indicates that graphical glitches frequently occur during the GUI rendering and severely degrade the quality of graphically-rich applications such as video games. Existing automated testing techniques for such applications focus mainly on generating various GUI test sequences and check whether the test sequences can cause crashes. These techniques require constant human attention to captures non-crashing bugs such as bugs causing graphical glitches. In this paper, we present the first step in automating the test oracle for detecting non-crashing bugs in graphically-rich applications. Specifically, we propose \texttt{GLIB} based on a code-based data augmentation technique to detect game GUI glitches. We perform an evaluation of \texttt{GLIB} on 20 real-world game apps (with bug reports available) and the result shows that \texttt{GLIB} can achieve 100\% precision and 99.5\% recall in detecting non-crashing bugs such as game GUI glitches. Practical application of \texttt{GLIB} on another 14 real-world games (without bug reports) further demonstrates that \texttt{GLIB} can effectively uncover GUI glitches, with 48 of 53 bugs reported by \texttt{GLIB} having been confirmed and fixed so far.

Xiangyu Liu, Hangtian Jia, Ying Wen et al.

Measuring and promoting policy diversity is critical for solving games with strong non-transitive dynamics where strategic cycles exist, and there is no consistent winner (e.g., Rock-Paper-Scissors). With that in mind, maintaining a pool of diverse policies via open-ended learning is an attractive solution, which can generate auto-curricula to avoid being exploited. However, in conventional open-ended learning algorithms, there are no widely accepted definitions for diversity, making it hard to construct and evaluate the diverse policies. In this work, we summarize previous concepts of diversity and work towards offering a unified measure of diversity in multi-agent open-ended learning to include all elements in Markov games, based on both Behavioral Diversity (BD) and Response Diversity (RD). At the trajectory distribution level, we re-define BD in the state-action space as the discrepancies of occupancy measures. For the reward dynamics, we propose RD to characterize diversity through the responses of policies when encountering different opponents. We also show that many current diversity measures fall in one of the categories of BD or RD but not both. With this unified diversity measure, we design the corresponding diversity-promoting objective and population effectivity when seeking the best responses in open-ended learning. We validate our methods in both relatively simple games like matrix game, non-transitive mixture model, and the complex \textit{Google Research Football} environment. The population found by our methods reveals the lowest exploitability, highest population effectivity in matrix game and non-transitive mixture model, as well as the largest goal difference when interacting with opponents of various levels in \textit{Google Research Football}.

Yu Chen, Yingfeng Chen, Zhipeng Hu et al.

Transfer learning (TL) is a promising way to improve the sample efficiency of reinforcement learning. However, how to efficiently transfer knowledge across tasks with different state-action spaces is investigated at an early stage. Most previous studies only addressed the inconsistency across different state spaces by learning a common feature space, without considering that similar actions in different action spaces of related tasks share similar semantics. In this paper, we propose a method to learning action embeddings by leveraging this idea, and a framework that learns both state embeddings and action embeddings to transfer policy across tasks with different state and action spaces. Our experimental results on various tasks show that the proposed method can not only learn informative action embeddings but accelerate policy learning.