Deep learning has become the standard approach for most machine learning tasks. While its impact is undeniable, interpreting the predictions of deep learning models from a human perspective remains a challenge. In contrast to model training, model interpretability is harder to quantify and pose as an explicit optimization problem. Inspired by the AUC softmax information curve (AUC SIC) metric for evaluating feature attribution methods, we propose a unified discrete optimization framework for feature attribution and feature selection based on subset selection. This leads to a natural adaptive generalization of the path integrated gradients (PIG) method for feature attribution, which we call Greedy PIG. We demonstrate the success of Greedy PIG on a wide variety of tasks, including image feature attribution, graph compression/explanation, and post-hoc feature selection on tabular data. Our results show that introducing adaptivity is a powerful and versatile method for making attribution methods more powerful.
Tracking dialogue states is an essential topic in task-oriented dialogue systems, which involve filling in the necessary information in pre-defined slots corresponding to a schema. While general pre-trained language models have been shown effective in slot-filling, their performance is limited when applied to specific domains. We propose a graph-based framework that learns domain-specific prompts by incorporating the dialogue schema. Specifically, we embed domain-specific schema encoded by a graph neural network into the pre-trained language model, which allows for relations in the schema to guide the model for better adaptation to the specific domain. Our experiments demonstrate that the proposed graph-based method outperforms other multi-domain DST approaches while using similar or fewer trainable parameters. We also conduct a comprehensive study of schema graph architectures, parameter usage, and module ablation that demonstrate the effectiveness of our model on multi-domain dialogue state tracking.
In recent years, audio-driven 3D facial animation has gained significant attention, particularly in applications such as virtual reality, gaming, and video conferencing. However, accurately modeling the intricate and subtle dynamics of facial expressions remains a challenge. Most existing studies approach the facial animation task as a single regression problem, which often fail to capture the intrinsic inter-modal relationship between speech signals and 3D facial animation and overlook their inherent consistency. Moreover, due to the limited availability of 3D-audio-visual datasets, approaches learning with small-size samples have poor generalizability that decreases the performance. To address these issues, in this study, we propose a cross-modal dual-learning framework, termed DualTalker, aiming at improving data usage efficiency as well as relating cross-modal dependencies. The framework is trained jointly with the primary task (audio-driven facial animation) and its dual task (lip reading) and shares common audio/motion encoder components. Our joint training framework facilitates more efficient data usage by leveraging information from both tasks and explicitly capitalizing on the complementary relationship between facial motion and audio to improve performance. Furthermore, we introduce an auxiliary cross-modal consistency loss to mitigate the potential over-smoothing underlying the cross-modal complementary representations, enhancing the mapping of subtle facial expression dynamics. Through extensive experiments and a perceptual user study conducted on the VOCA and BIWI datasets, we demonstrate that our approach outperforms current state-of-the-art methods both qualitatively and quantitatively. We have made our code and video demonstrations available at https://github.com/sabrina-su/iadf.git.
Effective training of deep image segmentation models is challenging due to the need for abundant, high-quality annotations. Generating annotations is laborious and time-consuming for human experts, especially in medical image segmentation. To facilitate image annotation, we introduce Physics Informed Contour Selection (PICS) - an interpretable, physics-informed algorithm for rapid image segmentation without relying on labeled data. PICS draws inspiration from physics-informed neural networks (PINNs) and an active contour model called snake. It is fast and computationally lightweight because it employs cubic splines instead of a deep neural network as a basis function. Its training parameters are physically interpretable because they directly represent control knots of the segmentation curve. Traditional snakes involve minimization of the edge-based loss functionals by deriving the Euler-Lagrange equation followed by its numerical solution. However, PICS directly minimizes the loss functional, bypassing the Euler Lagrange equations. It is the first snake variant to minimize a region-based loss function instead of traditional edge-based loss functions. PICS uniquely models the three-dimensional (3D) segmentation process with an unsteady partial differential equation (PDE), which allows accelerated segmentation via transfer learning. To demonstrate its effectiveness, we apply PICS for 3D segmentation of the left ventricle on a publicly available cardiac dataset. While doing so, we also introduce a new convexity-preserving loss term that encodes the shape information of the left ventricle to enhance PICS's segmentation quality. Overall, PICS presents several novelties in network architecture, transfer learning, and physics-inspired losses for image segmentation, thereby showing promising outcomes and potential for further refinement.
The automatic identification and analysis of pronunciation errors, known as Mispronunciation Detection and Diagnosis (MDD) plays a crucial role in Computer Aided Pronunciation Learning (CAPL) tools such as Second-Language (L2) learning or speech therapy applications. Existing MDD methods relying on analysing phonemes can only detect categorical errors of phonemes that have an adequate amount of training data to be modelled. With the unpredictable nature of the pronunciation errors of non-native or disordered speakers and the scarcity of training datasets, it is unfeasible to model all types of mispronunciations. Moreover, phoneme-level MDD approaches have a limited ability to provide detailed diagnostic information about the error made. In this paper, we propose a low-level MDD approach based on the detection of speech attribute features. Speech attribute features break down phoneme production into elementary components that are directly related to the articulatory system leading to more formative feedback to the learner. We further propose a multi-label variant of the Connectionist Temporal Classification (CTC) approach to jointly model the non-mutually exclusive speech attributes using a single model. The pre-trained wav2vec2 model was employed as a core model for the speech attribute detector. The proposed method was applied to L2 speech corpora collected from English learners from different native languages. The proposed speech attribute MDD method was further compared to the traditional phoneme-level MDD and achieved a significantly lower False Acceptance Rate (FAR), False Rejection Rate (FRR), and Diagnostic Error Rate (DER) over all speech attributes compared to the phoneme-level equivalent.
We present SPHINX, a versatile multi-modal large language model (MLLM) with a joint mixing of model weights, tuning tasks, and visual embeddings. First, for stronger vision-language alignment, we unfreeze the large language model (LLM) during pre-training, and introduce a weight mix strategy between LLMs trained by real-world and synthetic data. By directly integrating the weights from two domains, the mixed LLM can efficiently incorporate diverse semantics with favorable robustness. Then, to enable multi-purpose capabilities, we mix a variety of tasks for joint visual instruction tuning, and design task-specific instructions to avoid inter-task conflict. In addition to the basic visual question answering, we include more challenging tasks such as region-level understanding, caption grounding, document layout detection, and human pose estimation, contributing to mutual enhancement over different scenarios. Additionally, we propose to extract comprehensive visual embeddings from various network architectures, pre-training paradigms, and information granularity, providing language models with more robust image representations. Based on our proposed joint mixing, SPHINX exhibits superior multi-modal understanding capabilities on a wide range of applications. On top of this, we further propose an efficient strategy aiming to better capture fine-grained appearances of high-resolution images. With a mixing of different scales and high-resolution sub-images, SPHINX attains exceptional visual parsing and reasoning performance on existing evaluation benchmarks. We hope our work may cast a light on the exploration of joint mixing in future MLLM research. Code is released at https://github.com/Alpha-VLLM/LLaMA2-Accessory.
Text-based Person Search (TBPS) aims to retrieve images of target pedestrian indicated by textual descriptions. It is essential for TBPS to extract fine-grained local features and align them crossing modality. Existing methods utilize external tools or heavy cross-modal interaction to achieve explicit alignment of cross-modal fine-grained features, which is inefficient and time-consuming. In this work, we propose a Vision-Guided Semantic-Group Network (VGSG) for text-based person search to extract well-aligned fine-grained visual and textual features. In the proposed VGSG, we develop a Semantic-Group Textual Learning (SGTL) module and a Vision-guided Knowledge Transfer (VGKT) module to extract textual local features under the guidance of visual local clues. In SGTL, in order to obtain the local textual representation, we group textual features from the channel dimension based on the semantic cues of language expression, which encourages similar semantic patterns to be grouped implicitly without external tools. In VGKT, a vision-guided attention is employed to extract visual-related textual features, which are inherently aligned with visual cues and termed vision-guided textual features. Furthermore, we design a relational knowledge transfer, including a vision-language similarity transfer and a class probability transfer, to adaptively propagate information of the vision-guided textual features to semantic-group textual features. With the help of relational knowledge transfer, VGKT is capable of aligning semantic-group textual features with corresponding visual features without external tools and complex pairwise interaction. Experimental results on two challenging benchmarks demonstrate its superiority over state-of-the-art methods.
Network clustering tackles the problem of identifying sets of nodes (communities) that have similar connection patterns. However, in many scenarios, nodes also have attributes that are correlated with the clustering structure. Thus, network information (edges) and node information (attributes) can be jointly leveraged to design high-performance clustering algorithms. Under a general model for the network and node attributes, this work establishes an information-theoretic criterion for the exact recovery of community labels and characterizes a phase transition determined by the Chernoff-Hellinger divergence of the model. The criterion shows how network and attribute information can be exchanged in order to have exact recovery (e.g., more reliable network information requires less reliable attribute information). This work also presents an iterative clustering algorithm that maximizes the joint likelihood, assuming that the probability distribution of network interactions and node attributes belong to exponential families. This covers a broad range of possible interactions (e.g., edges with weights) and attributes (e.g., non-Gaussian models), as well as sparse networks, while also exploring the connection between exponential families and Bregman divergences. Extensive numerical experiments using synthetic data indicate that the proposed algorithm outperforms classic algorithms that leverage only network or only attribute information as well as state-of-the-art algorithms that also leverage both sources of information. The contributions of this work provide insights into the fundamental limits and practical techniques for inferring community labels on node-attributed networks.
Model-based RL is a promising approach for real-world robotics due to its improved sample efficiency and generalization capabilities compared to model-free RL. However, effective model-based RL solutions for vision-based real-world applications require bridging the sim-to-real gap for any world model learnt. Due to its significant computational cost, standard domain randomisation does not provide an effective solution to this problem. This paper proposes TWIST (Teacher-Student World Model Distillation for Sim-to-Real Transfer) to achieve efficient sim-to-real transfer of vision-based model-based RL using distillation. Specifically, TWIST leverages state observations as readily accessible, privileged information commonly garnered from a simulator to significantly accelerate sim-to-real transfer. Specifically, a teacher world model is trained efficiently on state information. At the same time, a matching dataset is collected of domain-randomised image observations. The teacher world model then supervises a student world model that takes the domain-randomised image observations as input. By distilling the learned latent dynamics model from the teacher to the student model, TWIST achieves efficient and effective sim-to-real transfer for vision-based model-based RL tasks. Experiments in simulated and real robotics tasks demonstrate that our approach outperforms naive domain randomisation and model-free methods in terms of sample efficiency and task performance of sim-to-real transfer.
With the rapid development of deep learning, a variety of change detection methods based on deep learning have emerged in recent years. However, these methods usually require a large number of training samples to train the network model, so it is very expensive. In this paper, we introduce a completely unsupervised shallow convolutional neural network (USCNN) fusion approach for change detection. Firstly, the bi-temporal images are transformed into different feature spaces by using convolution kernels of different sizes to extract multi-scale information of the images. Secondly, the output features of bi-temporal images at the same convolution kernels are subtracted to obtain the corresponding difference images, and the difference feature images at the same scale are fused into one feature image by using 1 * 1 convolution layer. Finally, the output features of different scales are concatenated and a 1 * 1 convolution layer is used to fuse the multi-scale information of the image. The model parameters are obtained by a redesigned sparse function. Our model has three features: the entire training process is conducted in an unsupervised manner, the network architecture is shallow, and the objective function is sparse. Thus, it can be seen as a kind of lightweight network model. Experimental results on four real remote sensing datasets indicate the feasibility and effectiveness of the proposed approach.