In this work, following the intuition that adverbs describing scene-sequences are best identified by reasoning over high-level concepts of object-behavior, we propose the design of a new framework that reasons over object-behaviours extracted from raw-video-clips to recognize the clip's corresponding adverb-types. Importantly, while previous works for general scene adverb-recognition assume knowledge of the clips underlying action-types, our method is directly applicable in the more general problem setting where the action-type of a video-clip is unknown. Specifically, we propose a novel pipeline that extracts human-interpretable object-behaviour-facts from raw video clips and propose novel symbolic and transformer based reasoning methods that operate over these extracted facts to identify adverb-types. Experiment results demonstrate that our proposed methods perform favourably against the previous state-of-the-art. Additionally, to support efforts in symbolic video-processing, we release two new datasets of object-behaviour-facts extracted from raw video clips - the MSR-VTT-ASP and ActivityNet-ASP datasets.
相關內容
In this work, we propose a novel framework for achieving robotic autonomy in orchards. It consists of two key steps: perception and semantic mapping. In the perception step, we introduce a 3D detection method that accurately identifies objects directly on point cloud maps. In the semantic mapping step, we develop a mapping module that constructs a visibility graph map by incorporating object-level information and terrain analysis. By combining these two steps, our framework improves the autonomy of agricultural robots in orchard environments. The accurate detection of objects and the construction of a semantic map enable the robot to navigate autonomously, perform tasks such as fruit harvesting, and acquire actionable information for efficient agricultural production.
We propose to apply several gradient estimation techniques to enable the differentiation of programs with discrete randomness in High Energy Physics. Such programs are common in High Energy Physics due to the presence of branching processes and clustering-based analysis. Thus differentiating such programs can open the way for gradient based optimization in the context of detector design optimization, simulator tuning, or data analysis and reconstruction optimization. We discuss several possible gradient estimation strategies, including the recent Stochastic AD method, and compare them in simplified detector design experiments. In doing so we develop, to the best of our knowledge, the first fully differentiable branching program.
To enhance on-road environmental perception for autonomous driving, accurate and real-time analytics on high-resolution video frames generated from on-board cameras be-comes crucial. In this paper, we design a lightweight object location method based on class activation mapping (CAM) to rapidly capture the region of interest (RoI) boxes that contain driving safety related objects from on-board cameras, which can not only improve the inference accuracy of vision tasks, but also reduce the amount of transmitted data. Considering the limited on-board computation resources, the RoI boxes extracted from the raw image are offloaded to the edge for further processing. Considering both the dynamics of vehicle-to-edge communications and the limited edge resources, we propose an adaptive RoI box offloading algorithm to ensure prompt and accurate inference by adjusting the down-sampling rate of each box. Extensive experimental results on four high-resolution video streams demonstrate that our approach can effectively improve the overall accuracy by up to 16% and reduce the transmission demand by up to 49%, compared with other benchmarks.
Human emotion understanding is pivotal in making conversational technology mainstream. We view speech emotion understanding as a perception task which is a more realistic setting. With varying contexts (languages, demographics, etc.) different share of people perceive the same speech segment as a non-unanimous emotion. As part of the ACM Multimedia 2023 Computational Paralinguistics ChallengE (ComParE) in the EMotion Share track, we leverage their rich dataset of multilingual speakers and multi-label regression target of 'emotion share' or perception of that emotion. We demonstrate that the training scheme of different foundation models dictates their effectiveness for tasks beyond speech recognition, especially for non-semantic speech tasks like emotion understanding. This is a very complex task due to multilingual speakers, variability in the target labels, and inherent imbalance in the regression dataset. Our results show that HuBERT-Large with a self-attention-based light-weight sequence model provides 4.6% improvement over the reported baseline.
A prevalent practice in recommender systems consists in averaging item embeddings to represent users or higher-level concepts in the same embedding space. This paper investigates the relevance of such a practice. For this purpose, we propose an expected precision score, designed to measure the consistency of an average embedding relative to the items used for its construction. We subsequently analyze the mathematical expression of this score in a theoretical setting with specific assumptions, as well as its empirical behavior on real-world data from music streaming services. Our results emphasize that real-world averages are less consistent for recommendation, which paves the way for future research to better align real-world embeddings with assumptions from our theoretical setting.
Visual Prompting (VP) is an emerging and powerful technique that allows sample-efficient adaptation to downstream tasks by engineering a well-trained frozen source model. In this work, we explore the benefits of VP in constructing compelling neural network classifiers with differential privacy (DP). We explore and integrate VP into canonical DP training methods and demonstrate its simplicity and efficiency. In particular, we discover that VP in tandem with PATE, a state-of-the-art DP training method that leverages the knowledge transfer from an ensemble of teachers, achieves the state-of-the-art privacy-utility trade-off with minimum expenditure of privacy budget. Moreover, we conduct additional experiments on cross-domain image classification with a sufficient domain gap to further unveil the advantage of VP in DP. Lastly, we also conduct extensive ablation studies to validate the effectiveness and contribution of VP under DP consideration. Our code is available at (//github.com/EzzzLi/Prompt-PATE).
The outdoor navigation capabilities of ground robots have improved significantly in recent years, opening up new potential applications in a variety of settings. Cost-based representations of the environment are frequently used in the path planning domain to obtain an optimized path based on various objectives, such as traversal time or energy consumption. However, obtaining such cost representations is still cumbersome, particularly in outdoor settings with diverse terrain types and slope angles. In this paper, we address this problem by using a data-driven approach to develop a cost representation for various outdoor terrain types that supports two optimization objectives, namely energy consumption and traversal time. We train a supervised machine learning model whose inputs consists of extracted environment data along a path and whose outputs are the predicted energy consumption and traversal time. The model is based on a ResNet neural network architecture and trained using field-recorded data. The error of the proposed method on different types of terrain is within 11\% of the ground truth data. To show that it performs and generalizes better than currently existing approaches on various types of terrain, a comparison to a baseline method is made.
Knowledge-based visual question answering is a very challenging and widely concerned task. Previous methods adopts the implicit knowledge in large language models (LLM) to achieve excellent results, but we argue that existing methods may suffer from biasing understanding of the image and insufficient knowledge to solve the problem. In this paper, we propose PROOFREAD -PROmpting vision language model with knOwledge From laRgE lAnguage moDel, a novel, lightweight and efficient kowledge-based VQA framework, which make the vision language model and the large language model cooperate to give full play to their respective strengths and bootstrap each other. In detail, our proposed method uses LLM to obtain knowledge explicitly, uses the vision language model which can see the image to get the knowledge answer, and introduces knowledge perceiver to filter out knowledge that is harmful for getting the correct final answer. Experimental results on two datasets prove the effectiveness of our approach. Our method outperforms all state-of-the-art methods on the A-OKVQA dataset in two settings and also achieves relatively good performance on the OKVQA dataset.
This work considers the question of how convenient access to copious data impacts our ability to learn causal effects and relations. In what ways is learning causality in the era of big data different from -- or the same as -- the traditional one? To answer this question, this survey provides a comprehensive and structured review of both traditional and frontier methods in learning causality and relations along with the connections between causality and machine learning. This work points out on a case-by-case basis how big data facilitates, complicates, or motivates each approach.
The previous work for event extraction has mainly focused on the predictions for event triggers and argument roles, treating entity mentions as being provided by human annotators. This is unrealistic as entity mentions are usually predicted by some existing toolkits whose errors might be propagated to the event trigger and argument role recognition. Few of the recent work has addressed this problem by jointly predicting entity mentions, event triggers and arguments. However, such work is limited to using discrete engineering features to represent contextual information for the individual tasks and their interactions. In this work, we propose a novel model to jointly perform predictions for entity mentions, event triggers and arguments based on the shared hidden representations from deep learning. The experiments demonstrate the benefits of the proposed method, leading to the state-of-the-art performance for event extraction.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191