Learning modality-fused representations and processing unaligned multimodal sequences are meaningful and challenging in multimodal emotion recognition. Existing approaches use directional pairwise attention or a message hub to fuse language, visual, and audio modalities. However, those approaches introduce information redundancy when fusing features and are inefficient without considering the complementarity of modalities. In this paper, we propose an efficient neural network to learn modality-fused representations with CB-Transformer (LMR-CBT) for multimodal emotion recognition from unaligned multimodal sequences. Specifically, we first perform feature extraction for the three modalities respectively to obtain the local structure of the sequences. Then, we design a novel transformer with cross-modal blocks (CB-Transformer) that enables complementary learning of different modalities, mainly divided into local temporal learning,cross-modal feature fusion and global self-attention representations. In addition, we splice the fused features with the original features to classify the emotions of the sequences. Finally, we conduct word-aligned and unaligned experiments on three challenging datasets, IEMOCAP, CMU-MOSI, and CMU-MOSEI. The experimental results show the superiority and efficiency of our proposed method in both settings. Compared with the mainstream methods, our approach reaches the state-of-the-art with a minimum number of parameters.
Recently, transformer-based image segmentation methods have achieved notable success against previous solutions. While for video domains, how to effectively model temporal context with the attention of object instances across frames remains an open problem. In this paper, we propose an online video instance segmentation framework with a novel instance-aware temporal fusion method. We first leverages the representation, i.e., a latent code in the global context (instance code) and CNN feature maps to represent instance- and pixel-level features. Based on this representation, we introduce a cropping-free temporal fusion approach to model the temporal consistency between video frames. Specifically, we encode global instance-specific information in the instance code and build up inter-frame contextual fusion with hybrid attentions between the instance codes and CNN feature maps. Inter-frame consistency between the instance codes are further enforced with order constraints. By leveraging the learned hybrid temporal consistency, we are able to directly retrieve and maintain instance identities across frames, eliminating the complicated frame-wise instance matching in prior methods. Extensive experiments have been conducted on popular VIS datasets, i.e. Youtube-VIS-19/21. Our model achieves the best performance among all online VIS methods. Notably, our model also eclipses all offline methods when using the ResNet-50 backbone.
As the volume of long-form spoken-word content such as podcasts explodes, many platforms desire to present short, meaningful, and logically coherent segments extracted from the full content. Such segments can be consumed by users to sample content before diving in, as well as used by the platform to promote and recommend content. However, little published work is focused on the segmentation of spoken-word content, where the errors (noise) in transcripts generated by automatic speech recognition (ASR) services poses many challenges. Here we build a novel dataset of complete transcriptions of over 400 podcast episodes, in which we label the position of introductions in each episode. These introductions contain information about the episodes' topics, hosts, and guests, providing a valuable summary of the episode content, as it is created by the authors. We further augment our dataset with word substitutions to increase the amount of available training data. We train three Transformer models based on the pre-trained BERT and different augmentation strategies, which achieve significantly better performance compared with a static embedding model, showing that it is possible to capture generalized, larger-scale structural information from noisy, loosely-organized speech data. This is further demonstrated through an analysis of the models' inner architecture. Our methods and dataset can be used to facilitate future work on the structure-based segmentation of spoken-word content.
Aerial robots are increasingly being utilized for a wide range of environmental monitoring and exploration tasks. However, a key challenge is efficiently planning paths to maximize the information value of acquired data as an initially unknown environment is explored. To address this, we propose a new approach for informative path planning (IPP) based on deep reinforcement learning (RL). Bridging the gap between recent advances in RL and robotic applications, our method combines Monte Carlo tree search with an offline-learned neural network predicting informative sensing actions. We introduce several components making our approach applicable for robotic tasks with continuous high-dimensional state spaces and large action spaces. By deploying the trained network during a mission, our method enables sample-efficient online replanning on physical platforms with limited computational resources. Evaluations using synthetic data show that our approach performs on par with existing information-gathering methods while reducing runtime by a factor of 8-10. We validate the performance of our framework using real-world surface temperature data from a crop field.
My doctoral research focuses on understanding semantic knowledge in neural network models trained solely to predict natural language (referred to as language models, or LMs), by drawing on insights from the study of concepts and categories grounded in cognitive science. I propose a framework inspired by 'inductive reasoning,' a phenomenon that sheds light on how humans utilize background knowledge to make inductive leaps and generalize from new pieces of information about concepts and their properties. Drawing from experiments that study inductive reasoning, I propose to analyze semantic inductive generalization in LMs using phenomena observed in human-induction literature, investigate inductive behavior on tasks such as implicit reasoning and emergent feature recognition, and analyze and relate induction dynamics to the learned conceptual representation space.
Recent research has shown that mixed-initiative conversational search, based on the interaction between users and computers to clarify and improve a query, provides enormous advantages. Nonetheless, incorporating additional information provided by the user from the conversation poses some challenges. In fact, further interactions could confuse the system as a user might use words irrelevant to the information need but crucial for correct sentence construction in the context of multi-turn conversations. To this aim, in this paper, we have collected two conversational keyword extraction datasets and propose an end-to-end document retrieval pipeline incorporating them. Furthermore, we study the performance of two neural keyword extraction models, namely, BERT and sequence to sequence, in terms of extraction accuracy and human annotation. Finally, we study the effect of keyword extraction on the end-to-end neural IR performance and show that our approach beats state-of-the-art IR models. We make the two datasets publicly available to foster research in this area.
Artificial intelligence has undergone immense growth and maturation in recent years, though autonomous systems have traditionally struggled when fielded in diverse and previously unknown environments. DARPA is seeking to change that with the Subterranean Challenge, by providing roboticists the opportunity to support civilian and military first responders in complex and high-risk underground scenarios. The subterranean domain presents a handful of challenges, such as limited communication, diverse topology and terrain, and degraded sensing. Team MARBLE proposes a solution for autonomous exploration of unknown subterranean environments in which coordinated agents search for artifacts of interest. The team presents two navigation algorithms in the form of a metric-topological graph-based planner and a continuous frontier-based planner. To facilitate multi-agent coordination, agents share and merge new map information and candidate goal-points. Agents deploy communication beacons at different points in the environment, extending the range at which maps and other information can be shared. Onboard autonomy reduces the load on human supervisors, allowing agents to detect and localize artifacts and explore autonomously outside established communication networks. Given the scale, complexity, and tempo of this challenge, a range of lessons were learned, most importantly, that frequent and comprehensive field testing in representative environments is key to rapidly refining system performance.
In modern society, digital images have become a prominent source of information and medium of communication. They can, however, be simply altered using freely available image editing software. Two or more images are combined to generate a new image that can transmit information across social media platforms to influence the people in the society. This information may have both positive and negative consequences. Hence there is a need to develop a technique that will detect and locates a multiple image splicing forgery in an image. This research work proposes multiple image splicing forgery detection using Mask R-CNN, with a backbone as a MobileNet V1. It also calculates the percentage score of a forged region of multiple spliced images. The comparative analysis of the proposed work with the variants of ResNet is performed. The proposed model is trained and tested using our MISD (Multiple Image Splicing Dataset), and it is observed that the proposed model outperforms the variants of ResNet models (ResNet 51,101 and 151).
We develop a model of social learning from overabundant information: Short-lived agents sequentially choose from a large set of (flexibly correlated) information sources for prediction of an unknown state. Signal realizations are public. We demonstrate two starkly different long-run outcomes: (1) efficient information aggregation, where the community eventually learns as fast as possible; (2) "learning traps," where the community gets stuck observing suboptimal sources and learns inefficiently. Our main results identify a simple property of the signal correlation structure that separates these outcomes. In both regimes, we characterize which sources are observed in the long run and how often.
This paper studies the BERT pretraining of video transformers. It is a straightforward but worth-studying extension given the recent success from BERT pretraining of image transformers. We introduce BEVT which decouples video representation learning into spatial representation learning and temporal dynamics learning. In particular, BEVT first performs masked image modeling on image data, and then conducts masked image modeling jointly with masked video modeling on video data. This design is motivated by two observations: 1) transformers learned on image datasets provide decent spatial priors that can ease the learning of video transformers, which are often times computationally-intensive if trained from scratch; 2) discriminative clues, i.e., spatial and temporal information, needed to make correct predictions vary among different videos due to large intra-class and inter-class variations. We conduct extensive experiments on three challenging video benchmarks where BEVT achieves very promising results. On Kinetics 400, for which recognition mostly relies on discriminative spatial representations, BEVT achieves comparable results to strong supervised baselines. On Something-Something-V2 and Diving 48, which contain videos relying on temporal dynamics, BEVT outperforms by clear margins all alternative baselines and achieves state-of-the-art performance with a 70.6% and 86.7% Top-1 accuracy respectively.