We consider the problem of sequential sparse subset selections in an online learning setup. Assume that the set $[N]$ consists of $N$ distinct elements. On the $t^{\text{th}}$ round, a monotone reward function $f_t: 2^{[N]} \to \mathbb{R}_+,$ which assigns a non-negative reward to each subset of $[N],$ is revealed to a learner. The learner selects (perhaps randomly) a subset $S_t \subseteq [N]$ of $k$ elements before the reward function $f_t$ for that round is revealed $(k \leq N)$. As a consequence of its choice, the learner receives a reward of $f_t(S_t)$ on the $t^{\text{th}}$ round. The learner's goal is to design an online subset selection policy to maximize its expected cumulative reward accrued over a given time horizon. In this connection, we propose an online learning policy called SCore (Subset Selection with Core) that solves the problem for a large class of reward functions. The proposed SCore policy is based on a new concept of $\alpha$-Core, which is a generalization of the notion of Core from the cooperative game theory literature. We establish a learning guarantee for the SCore policy in terms of a new performance metric called $\alpha$-augmented regret. In this new metric, the power of the offline benchmark is suitably augmented compared to the online policy. We give several illustrative examples to show that a broad class of reward functions, including submodular, can be efficiently learned using the SCore policy. We also outline how the SCore policy can be used under a semi-bandit feedback model and conclude the paper with a number of open problems.
Compiling comprehensive repositories of commonsense knowledge is a long-standing problem in AI. Many concerns revolve around the issue of reporting bias, i.e., that frequency in text sources is not a good proxy for relevance or truth. This paper explores whether children's texts hold the key to commonsense knowledge compilation, based on the hypothesis that such content makes fewer assumptions on the reader's knowledge, and therefore spells out commonsense more explicitly. An analysis with several corpora shows that children's texts indeed contain much more, and more typical commonsense assertions. Moreover, experiments show that this advantage can be leveraged in popular language-model-based commonsense knowledge extraction settings, where task-unspecific fine-tuning on small amounts of children texts (childBERT) already yields significant improvements. This provides a refreshing perspective different from the common trend of deriving progress from ever larger models and corpora.
Recent text-to-image matching models apply contrastive learning to large corpora of uncurated pairs of images and sentences. While such models can provide a powerful score for matching and subsequent zero-shot tasks, they are not capable of generating caption given an image. In this work, we repurpose such models to generate a descriptive text given an image at inference time, without any further training or tuning step. This is done by combining the visual-semantic model with a large language model, benefiting from the knowledge in both web-scale models. The resulting captions are much less restrictive than those obtained by supervised captioning methods. Moreover, as a zero-shot learning method, it is extremely flexible and we demonstrate its ability to perform image arithmetic in which the inputs can be either images or text and the output is a sentence. This enables novel high-level vision capabilities such as comparing two images or solving visual analogy tests.
We propose an accurate and efficient scene text detection framework, termed FAST (i.e., faster arbitrarily-shaped text detector). Different from recent advanced text detectors that used hand-crafted network architectures and complicated post-processing, resulting in low inference speed, FAST has two new designs. (1) We search the network architecture by designing a network search space and reward function carefully tailored for text detection, leading to more powerful features than most networks that are searched for image classification. (2) We design a minimalist representation (only has 1-channel output) to model text with arbitrary shape, as well as a GPU-parallel post-processing to efficiently assemble text lines with negligible time overhead. Benefiting from these two designs, FAST achieves an excellent trade-off between accuracy and efficiency on several challenging datasets. For example, FAST-A0 yields 81.4% F-measure at 152 FPS on Total-Text, outperforming the previous fastest method by 1.5 points and 70 FPS in terms of accuracy and speed. With TensorRT optimization, the inference speed can be further accelerated to over 600 FPS.
Knowledge-enhanced methods that take advantage of auxiliary knowledge graphs recently emerged in relation extraction, and they surpass traditional text-based relation extraction methods. However, there are no unified public benchmarks that currently involve evidence sentences and knowledge graphs for knowledge-enhanced relation extraction. To combat these issues, we propose KGRED, a knowledge graph enhanced relation extraction dataset with features as follows: (1) the benchmarks are based on widely-used distantly supervised relation extraction datasets; (2) we refine these existing datasets to improve the data quality, and we also construct auxiliary knowledge graphs for these existing datasets through entity linking to support knowledge-enhanced relation extraction tasks; (3) with the new benchmarks we curated, we build baselines in two popular relation extraction settings including sentence-level and bag-level relation extraction, and we also make comparisons among the latest knowledge-enhanced relation extraction methods. KGRED provides high-quality relation extraction datasets with auxiliary knowledge graphs for evaluating the performance of knowledge-enhanced relation extraction methods. Meanwhile, our experiments on KGRED reveal the influence of knowledge graph information on relation extraction tasks.
Named entity recognition is a fundamental task in natural language processing, identifying the span and category of entities in unstructured texts. The traditional sequence labeling methodology ignores the nested entities, i.e. entities included in other entity mentions. Many approaches attempt to address this scenario, most of which rely on complex structures or have high computation complexity. The representation learning of the heterogeneous star graph containing text nodes and type nodes is investigated in this paper. In addition, we revise the graph attention mechanism into a hybrid form to address its unreasonableness in specific topologies. The model performs the type-supervised sequence labeling after updating nodes in the graph. The annotation scheme is an extension of the single-layer sequence labeling and is able to cope with the vast majority of nested entities. Extensive experiments on public NER datasets reveal the effectiveness of our model in extracting both flat and nested entities. The method achieved state-of-the-art performance on both flat and nested datasets. The significant improvement in accuracy reflects the superiority of the multi-layer labeling strategy.
Generative models are becoming popular for the synthesis of medical images. Recently, neural diffusion models have demonstrated the potential to generate photo-realistic images of objects. However, their potential to generate medical images is not explored yet. In this work, we explore the possibilities of synthesis of medical images using neural diffusion models. First, we use a pre-trained DALLE2 model to generate lungs X-Ray and CT images from an input text prompt. Second, we train a stable diffusion model with 3165 X-Ray images and generate synthetic images. We evaluate the synthetic image data through a qualitative analysis where two independent radiologists label randomly chosen samples from the generated data as real, fake, or unsure. Results demonstrate that images generated with the diffusion model can translate characteristics that are otherwise very specific to certain medical conditions in chest X-Ray or CT images. Careful tuning of the model can be very promising. To the best of our knowledge, this is the first attempt to generate lungs X-Ray and CT images using neural diffusion models. This work aims to introduce a new dimension in artificial intelligence for medical imaging. Given that this is a new topic, the paper will serve as an introduction and motivation for the research community to explore the potential of diffusion models for medical image synthesis. We have released the synthetic images on https://www.kaggle.com/datasets/hazrat/awesomelungs.
Multi-label image classification is a foundational topic in various domains. Multimodal learning approaches have recently achieved outstanding results in image representation and single-label image classification. For instance, Contrastive Language-Image Pretraining (CLIP) demonstrates impressive image-text representation learning abilities and is robust to natural distribution shifts. This success inspires us to leverage multimodal learning for multi-label classification tasks, and benefit from contrastively learnt pretrained models. We propose the Multimodal Multi-label Image Classification (MuMIC) framework, which utilizes a hardness-aware tempered sigmoid based Binary Cross Entropy loss function, thus enables the optimization on multi-label objectives and transfer learning on CLIP. MuMIC is capable of providing high classification performance, handling real-world noisy data, supporting zero-shot predictions, and producing domain-specific image embeddings. In this study, a total of 120 image classes are defined, and more than 140K positive annotations are collected on approximately 60K Booking.com images. The final MuMIC model is deployed on Booking.com Content Intelligence Platform, and it outperforms other state-of-the-art models with 85.6% GAP@10 and 83.8% GAP on all 120 classes, as well as a 90.1% macro mAP score across 32 majority classes. We summarize the modeling choices which are extensively tested through ablation studies. To the best of our knowledge, we are the first to adapt contrastively learnt multimodal pretraining for real-world multi-label image classification problems, and the innovation can be transferred to other domains.
While recent studies have focused on quantifying word usage to find the overall shapes of narrative emotional arcs, certain features of narratives within narratives remain to be explored. Here, we characterize the narrative time scale of sub-narratives by finding the length of text at which fluctuations in word usage begin to be relevant. We represent more than 30,000 Project Gutenberg books as time series using ousiometrics, a power-danger framework for essential meaning, itself a reinterpretation of the valence-arousal-dominance framework derived from semantic differentials. We decompose each book's power and danger time series using empirical mode decomposition into a sum of constituent oscillatory modes and a non-oscillatory trend. By comparing the decomposition of the original power and danger time series with those derived from shuffled text, we find that shorter books exhibit only a general trend, while longer books have fluctuations in addition to the general trend, similar to how subplots have arcs within an overall narrative arc. These fluctuations typically have a period of a few thousand words regardless of the book length or library classification code, but vary depending on the content and structure of the book. Our method provides a data-driven denoising approach that works for text of various lengths, in contrast to the more traditional approach of using large window sizes that may inadvertently smooth out relevant information, especially for shorter texts.
Current NLP techniques have been greatly applied in different domains. In this paper, we propose a human-in-the-loop framework for robotic grasping in cluttered scenes, investigating a language interface to the grasping process, which allows the user to intervene by natural language commands. This framework is constructed on a state-of-the-art rasping baseline, where we substitute a scene-graph representation with a text representation of the scene using BERT. Experiments on both simulation and physical robot show that the proposed method outperforms conventional object-agnostic and scene-graph based methods in the literature. In addition, we find that with human intervention, performance can be significantly improved.