In this work, we tackle action-scene hallucination in Video Large Language Models (Video-LLMs), where models incorrectly predict actions based on the scene context or scenes based on observed actions. We observe that existing Video-LLMs often suffer from action-scene hallucination due to two main factors. First, existing Video-LLMs intermingle spatial and temporal features by applying an attention operation across all tokens. Second, they use standard Rotary Position Embedding (RoPE), which causes the text tokens to overemphasize certain types of tokens depending on their sequential orders. To address these issues, we introduce MASH-VLM, Mitigating Action-Scene Hallucination in Video-LLMs through disentangled spatial-temporal representations. Our approach includes two key innovations: (1) DST-attention, a novel attention mechanism that disentangles spatial and temporal tokens within the LLM by using masked attention to restrict direct interactions be tween spatial and temporal tokens; (2) Harmonic-RoPE, which extends the dimensionality of the positional IDs, allowing spatial and temporal tokens to maintain balanced positions relative to the text tokens. To evaluate the action-scene hallucination in Video-LLMs, we introduce the UN-SCENE benchmark with 1,320 videos and 4,078 QA pairs. MASH-VLM achieves state-of-the-art performance on the UNSCENE benchmark, as well as on existing video understanding benchmarks.

Recent advances in vision language pretraining (VLP) have been largely attributed to the large-scale data collected from the web. However, uncurated dataset contains weakly correlated image-text pairs, causing data inefficiency. To address the issue, knowledge distillation have been explored at the expense of extra image and text momentum encoders to generate teaching signals for misaligned image-text pairs. In this paper, our goal is to resolve the misalignment problem with an efficient distillation framework. To this end, we propose ECLIPSE: Expediting Contrastive Language-Image Pretraining with Self-distilled Encoders. ECLIPSE features a distinctive distillation architecture wherein a shared text encoder is utilized between an online image encoder and a momentum image encoder. This strategic design choice enables the distillation to operate within a unified projected space of text embedding, resulting in better performance. Based on the unified text embedding space, ECLIPSE compensates for the additional computational cost of the momentum image encoder by expediting the online image encoder. Through our extensive experiments, we validate that there is a sweet spot between expedition and distillation where the partial view from the expedited online image encoder interacts complementarily with the momentum teacher. As a result, ECLIPSE outperforms its counterparts while achieving substantial acceleration in inference speed.

Efficiency and performance are significant challenges in applying Machine Learning (ML) to robotics, especially in energy-constrained real-world scenarios. In this context, Hyperdimensional Computing offers an energy-efficient alternative but has been underexplored in robotics. We introduce ReactHD, an HDC-based framework tailored for perception-action-based learning for sensorimotor controls of robot tasks. ReactHD employs hypervectors to encode sensory inputs and learn the suitable high-dimensional pattern for robot actions. It also integrates two HD-based lightweight symbolic learning techniques: HDC-based supervised learning by demonstration (HDC-IL) and HD-Reinforcement Learning (HDC-RL) to enable precise, reactive robot behaviors in complex environments. Our empirical evaluations show that ReactHD achieves robust and accurate learning outcomes comparable to state-of-the-art deep learning while substantially improving the performance and energy consumption efficiency by 14.2× and 15.3×. To the best of our knowledge, ReactHD is the first HDC-based framework deployed in real-world settings.

Recent self-supervised video representation learning methods focus on maximizing the similarity between multiple augmented views from the same video and largely rely on the quality of generated views. However, most existing methods lack a mechanism to prevent representation learning from bias towards static information in the video. In this paper, we propose frequency augmentation (FreqAug), a spatio-temporal data augmentation method in the frequency domain for video representation learning. FreqAug stochastically removes specific frequency components from the video so that learned representation captures essential features more from the remaining information for various downstream tasks. Specifically, FreqAug pushes the model to focus more on dynamic features rather than static features in the video via dropping spatial or temporal low-frequency components. To verify the generality of the proposed method, we experiment with FreqAug on multiple self-supervised learning frameworks along with standard augmentations. Transferring the improved representation to five video action recognition and two temporal action localization downstream tasks shows consistent improvements over baselines.

Contrastive Language-Image Pretraining has emerged as a prominent approach for training vision and text encoders with uncurated image-text pairs from the web. To enhance data-efficiency, recent efforts have introduced additional supervision terms that involve random-augmented views of the image. However, since the image augmentation process is unaware of its text counterpart, this procedure could cause various degrees of image-text misalignments during training. Prior methods either disregarded this discrepancy or introduced external models to mitigate the impact of misalignments during training. In contrast, we propose a novel metric learning approach that capitalizes on these misalignments as an additional training source, which we term "Misalign, Contrast then Distill (MCD)". Unlike previous methods that treat augmented images and their text counterparts as simple positive pairs, MCD predicts the continuous scales of misalignment caused by the augmentation. Our extensive experimental results show that our proposed MCD achieves state-of-the-art transferability in multiple classification and retrieval downstream datasets.

Deep neural networks for video action recognition frequently require 3D convolutional filters and often encounter overfitting due to a larger number of parameters. In this paper, we propose Random Mean Scaling (RMS), a simple and effective regularization method, to relieve the overfitting problem in 3D residual networks. The key idea of RMS is to randomly vary the magnitude of low-frequency components of the feature to regularize the model. The low-frequency component can be derived by a spatio-temporal mean on the local patch of a feature. We present that selective regularization on this locally smoothed feature makes a model handle the low-frequency and high-frequency component distinctively, resulting in performance improvement. RMS can enhance a model with little additional computation only during training, similar to other regularization methods. RMS also can be incorporated into typical training process without any bells and whistles. Experimental results show the improvement in generalization performance on a popular action recognition datasets demonstrating the effectiveness of RMS as a regularization technique, compared to other state-of-the-art regularization methods.

This paper aims to investigate how adequate cognitive functions for recognizing, predicting, and generating a variety of actions can be developed through iterative learning of action-caused dynamic perceptual patterns. Particularly, we examined the capabilities of mental simulation of one's own actions as well as the inference of others' intention because they play a crucial role, especially in social cognition. We propose a dynamic neural network model based on predictive coding which can generate and recognize dynamic visuo-proprioceptive patterns. The proposed model was examined by conducting a set of robotic simulation experiments in which a robot was trained to imitate visually perceived gesture patterns of human subjects in a simulation environment. The experimental results showed that the proposed model was able to develop a predictive model of imitative interaction through iterative learning of large-scale spatio-temporal patterns in visuo-proprioceptive input streams. Also, the experiment verified that the model was able to generate mental imagery of dynamic visuo-proprioceptive patterns without feeding the external inputs. Furthermore, the model was able to recognize the intention of others by minimizing prediction error in the observations of the others' action patterns in an online manner. These findings suggest that the error minimization principle in predictive coding could provide a primal account for the mirror neuron functions for generating actions as well as recognizing those generated by others in a social cognitive context.

XML has grown to represent and exchange information which is used widely in various computing areas such as web environment, mobile environment, data management systems. Especially, the most important issue for practical purposes is how to achieve the interoperability between XML model and relational database model. Until now, various algorithms have been proposed to achieve it. However, existing algorithms do not consider implicit referential integrity relation. VP-T and QP-T have been proposed to resolve the problem — QP-T enhanced VP-T based on query pattern. However, both have a critical restriction that column titles for an attribute are identically represented. Therefore, several referential integrity relations might not be extracted. In this paper, we introduce a novel translation algorithm whose name is WordNet-based Translation algorithm (WB-T) to resolve the above issues. WB-T can check similarity between column titles based on WordNet and extract more exact implicit referential integrity relation than the previous algorithms. WB-T provides improved extraction accuracy and reduces the extraction time.

This paper proposes a semantic management framework providing the ubiquitous application fieldindependent interoperability. The goal of the proposed framework is to provide a consistent and independent semantic usability to ubiquitous applications. The proposed framework uses the merits of the metadata registry that has been developed for exchanging and sharing between databases by ISO/IEC JTC 1/SC 32. The proposed framework allows application field free semantic management and enables the practical and ideal ubiquitous application development.