Recently, prompt-based methods have emerged as a new alternative `parameter-efficient fine-tuning' paradigm, which only fine-tunes a small number of additional parameters while keeping the original model frozen. However, despite achieving notable results, existing prompt methods mainly focus on `what to add', while overlooking the equally important aspect of `where to add', typically relying on the manually crafted placement. To this end, we propose a region-based Adaptive Visual Prompt, named AdaViPro, which integrates the `where to add' optimization of the prompt into the learning process. Specifically, we reconceptualize the `where to add' optimization as a problem of regional decision-making. During inference, AdaViPro generates a regionalized mask map for the whole image, which is composed of 0 and 1, to designate whether to apply or discard the prompt in each specific area. Therefore, we employ Gumbel-Softmax sampling to enable AdaViPro's end-to-end learning through standard back-propagation. Extensive experiments demonstrate that our AdaViPro yields new efficiency and accuracy trade-offs for adapting pre-trained models.
Extremely large-scale multiple-input-multiple-output (XL-MIMO) is a promising technique to achieve ultra-high spectral efficiency for future 6G communications. The mixed line-of-sight (LoS) and non-line-of-sight (NLoS) XL-MIMO near-field channel model is adopted to describe the XL-MIMO near-field channel accurately. In this paper, a generative adversarial network (GAN) variant based channel estimation method is proposed for XL-MIMO systems. Specifically, the GAN variant is developed to simultaneously estimate the LoS and NLoS path components of the XL-MIMO channel. The initially estimated channels instead of the received signals are input into the GAN variant as the conditional input to generate the XL-MIMO channels more efficiently. The GAN variant not only learns the mapping from the initially estimated channels to the XL-MIMO channels but also learns an adversarial loss. Moreover, we combine the adversarial loss with a conventional loss function to ensure the correct direction of training the generator. To further enhance the estimation performance, we investigate the impact of the hyper-parameter of the loss function on the performance of our method. Simulation results show that the proposed method outperforms the existing channel estimation approaches in the adopted channel model. In addition, the proposed method surpasses the Cram$\acute{\mathbf{e}}$r-Rao lower bound (CRLB) under low pilot overhead.
This paper presents a deep learning enhanced adaptive unscented Kalman filter (UKF) for predicting human arm motion in the context of manufacturing. Unlike previous network-based methods that solely rely on captured human motion data, which is represented as bone vectors in this paper, we incorporate a human arm dynamic model into the motion prediction algorithm and use the UKF to iteratively forecast human arm motions. Specifically, a Lagrangian-mechanics-based physical model is employed to correlate arm motions with associated muscle forces. Then a Recurrent Neural Network (RNN) is integrated into the framework to predict future muscle forces, which are transferred back to future arm motions based on the dynamic model. Given the absence of measurement data for future human motions that can be input into the UKF to update the state, we integrate another RNN to directly predict human future motions and treat the prediction as surrogate measurement data fed into the UKF. A noteworthy aspect of this study involves the quantification of uncertainties associated with both the data-driven and physical models in one unified framework. These quantified uncertainties are used to dynamically adapt the measurement and process noises of the UKF over time. This adaption, driven by the uncertainties of the RNN models, addresses inaccuracies stemming from the data-driven model and mitigates discrepancies between the assumed and true physical models, ultimately enhancing the accuracy and robustness of our predictions. Compared to the traditional RNN-based prediction, our method demonstrates improved accuracy and robustness in extensive experimental validations of various types of human motions.
Video grounding aims to localize the target moment in an untrimmed video corresponding to a given sentence query. Existing methods typically select the best prediction from a set of predefined proposals or directly regress the target span in a single-shot manner, resulting in the absence of a systematical prediction refinement process. In this paper, we propose DiffusionVG, a novel framework with diffusion models that formulates video grounding as a conditional generation task, where the target span is generated from Gaussian noise inputs and interatively refined in the reverse diffusion process. During training, DiffusionVG progressively adds noise to the target span with a fixed forward diffusion process and learns to recover the target span in the reverse diffusion process. In inference, DiffusionVG can generate the target span from Gaussian noise inputs by the learned reverse diffusion process conditioned on the video-sentence representations. Our DiffusionVG follows the encoder-decoder architecture, which firstly encodes the video-sentence features and iteratively denoises the predicted spans in its specialized span refining decoder. Without bells and whistles, our DiffusionVG demonstrates competitive or even superior performance compared to existing well-crafted models on mainstream Charades-STA and ActivityNet Captions benchmarks.
Emotion recognition in conversations (ERC) is a rapidly evolving task within the natural language processing community, which aims to detect the emotions expressed by speakers during a conversation. Recently, a growing number of ERC methods have focused on leveraging supervised contrastive learning (SCL) to enhance the robustness and generalizability of learned features. However, current SCL-based approaches in ERC are impeded by the constraint of large batch sizes and the lack of compatibility with most existing ERC models. To address these challenges, we propose an efficient and model-agnostic SCL framework named Supervised Sample-Label Contrastive Learning with Soft-HGR Maximal Correlation (SSLCL), which eliminates the need for a large batch size and can be seamlessly integrated with existing ERC models without introducing any model-specific assumptions. Specifically, we introduce a novel perspective on utilizing label representations by projecting discrete labels into dense embeddings through a shallow multilayer perceptron, and formulate the training objective to maximize the similarity between sample features and their corresponding ground-truth label embeddings, while minimizing the similarity between sample features and label embeddings of disparate classes. Moreover, we innovatively adopt the Soft-HGR maximal correlation as a measure of similarity between sample features and label embeddings, leading to significant performance improvements over conventional similarity measures. Additionally, multimodal cues of utterances are effectively leveraged by SSLCL as data augmentations to boost model performances. Extensive experiments on two ERC benchmark datasets, IEMOCAP and MELD, demonstrate the compatibility and superiority of our proposed SSLCL framework compared to existing state-of-the-art SCL methods. Our code is available at \url{https://github.com/TaoShi1998/SSLCL}.
Product disassembly plays a crucial role in the recycling, remanufacturing, and reuse of end-of-use (EoU) products. However, the current manual disassembly process is inefficient due to the complexity and variation of EoU products. While fully automating disassembly is not economically viable given the intricate nature of the task, there is potential in using human-robot collaboration (HRC) to enhance disassembly operations. HRC combines the flexibility and problem-solving abilities of humans with the precise repetition and handling of unsafe tasks by robots. Nevertheless, numerous challenges persist in technology, human workers, and remanufacturing work, that require comprehensive multidisciplinary research to bridge critical gaps. These challenges have motivated the authors to provide a detailed discussion on the opportunities and obstacles associated with introducing HRC to disassembly. In this regard, the authors have conducted a thorough review of the recent progress in HRC disassembly and present the insights gained from this analysis from three distinct perspectives: technology, workers, and work.
Historical behaviors have shown great effect and potential in various prediction tasks, including recommendation and information retrieval. The overall historical behaviors are various but noisy while search behaviors are always sparse. Most existing approaches in personalized search ranking adopt the sparse search behaviors to learn representation with bottleneck, which do not sufficiently exploit the crucial long-term interest. In fact, there is no doubt that user long-term interest is various but noisy for instant search, and how to exploit it well still remains an open problem. To tackle this problem, in this work, we propose a novel model named Query-dominant user Interest Network (QIN), including two cascade units to filter the raw user behaviors and reweigh the behavior subsequences. Specifically, we propose a relevance search unit (RSU), which aims to search a subsequence relevant to the query first and then search the sub-subsequences relevant to the target item. These items are then fed into an attention unit called Fused Attention Unit (FAU). It should be able to calculate attention scores from the ID field and attribute field separately, and then adaptively fuse the item embedding and content embedding based on the user engagement of past period. Extensive experiments and ablation studies on real-world datasets demonstrate the superiority of our model over state-of-the-art methods. The QIN now has been successfully deployed on Kuaishou search, an online video search platform, and obtained 7.6% improvement on CTR.
Accurate deformable object manipulation (DOM) is essential for achieving autonomy in robotic surgery, where soft tissues are being displaced, stretched, and dissected. Many DOM methods can be powered by simulation, which ensures realistic deformation by adhering to the governing physical constraints and allowing for model prediction and control. However, real soft objects in robotic surgery, such as membranes and soft tissues, have complex, anisotropic physical parameters that a simulation with simple initialization from cameras may not fully capture. To use the simulation techniques in real surgical tasks, the "real-to-sim" gap needs to be properly compensated. In this work, we propose an online, adaptive parameter tuning approach for simulation optimization that (1) bridges the real-to-sim gap between a physics simulation and observations obtained 3D perceptions through estimating a residual mapping and (2) optimizes its stiffness parameters online. Our method ensures a small residual gap between the simulation and observation and improves the simulation's predictive capabilities. The effectiveness of the proposed mechanism is evaluated in the manipulation of both a thin-shell and volumetric tissue, representative of most tissue scenarios. This work contributes to the advancement of simulation-based deformable tissue manipulation and holds potential for improving surgical autonomy.
Cloth manipulation is a category of deformable object manipulation of great interest to the robotics community, from applications of automated laundry-folding and home organizing and cleaning to textiles and flexible manufacturing. Despite the desire for automated cloth manipulation, the thin-shell dynamics and under-actuation nature of cloth present significant challenges for robots to effectively interact with them. Many recent works omit explicit modeling in favor of learning-based methods that may yield control policies directly. However, these methods require large training sets that must be collected and curated. In this regard, we create a framework for differentiable modeling of cloth dynamics leveraging an Extended Position-based Dynamics (XPBD) algorithm. Together with the desired control objective, physics-aware regularization terms are designed for better results, including trajectory smoothness and elastic potential energy. In addition, safety constraints, such as avoiding obstacles, can be specified using signed distance functions (SDFs). We formulate the cloth manipulation task with safety constraints as a constrained optimization problem, which can be effectively solved by mainstream gradient-based optimizers thanks to the end-to-end differentiability of our framework. Finally, we assess the proposed framework for manipulation tasks with various safety thresholds and demonstrate the feasibility of result trajectories on a surgical robot. The effects of the regularization terms are analyzed in an additional ablation study.
Although dominant in natural language processing, transformer-based models remain challenged by the task of long-sequence processing, because the computational cost of self-attention operations in transformers swells quadratically with the input sequence length. To alleviate the complexity of long-sequence processing, we propose a simple framework to enable the offthe-shelf pre-trained transformers to process much longer sequences, while the computation and memory costs remain growing linearly with the input sequence lengths. More specifically, our method divides each long-sequence input into a batch of chunks, then aligns the interchunk information during the encoding steps, and finally selects the most representative hidden states from the encoder for the decoding process. To extract inter-chunk semantic information, we align the start and end token embeddings among chunks in each encoding transformer block. To learn an effective hidden selection policy, we design a dual updating scheme inspired by reinforcement learning, which regards the decoders of transformers as environments, and the downstream performance metrics as the rewards to evaluate the hidden selection actions. Our empirical results on real-world long-text summarization and reading comprehension tasks demonstrate effective improvements compared to prior longsequence processing baselines.