Unsupervised learning of keypoints and landmarks has seen significant progress with the help of modern neural network architectures, but performance is yet to match the supervised counterpart, making their practicability questionable. We leverage the emergent knowledge within text-to-image diffusion models, towards more robust unsupervised keypoints. Our core idea is to find text embeddings that would cause the generative model to consistently attend to compact regions in images (i.e. keypoints). To do so, we simply optimize the text embedding such that the cross-attention maps within the denoising network are localized as Gaussians with small standard deviations. We validate our performance on multiple datasets: the CelebA, CUB-200-2011, Tai-Chi-HD, DeepFashion, and Human3.6m datasets. We achieve significantly improved accuracy, sometimes even outperforming supervised ones, particularly for data that is non-aligned and less curated. Our code is publicly available and can be found through our project page: https://ubc-vision.github.io/StableKeypoints/
Driven by the upsurge progress in text-to-image (T2I) generation models, text-to-video (T2V) generation has experienced a significant advance as well. Accordingly, tasks such as modifying the object or changing the style in a video have been possible. However, previous works usually work well on trivial and consistent shapes, and easily collapse on a difficult target that has a largely different body shape from the original one. In this paper, we spot the bias problem in the existing video editing method that restricts the range of choices for the new protagonist and attempt to address this issue using the conventional image-level personalization method. We adopt motion personalization that isolates the motion from a single source video and then modifies the protagonist accordingly. To deal with the natural discrepancy between image and video, we propose a motion word with an inflated textual embedding to properly represent the motion in a source video. We also regulate the motion word to attend to proper motion-related areas by introducing a novel pseudo optical flow, efficiently computed from the pre-calculated attention maps. Finally, we decouple the motion from the appearance of the source video with an additional pseudo word. Extensive experiments demonstrate the editing capability of our method, taking a step toward more diverse and extensive video editing.
The application of Machine learning to finance has become a familiar approach, even more so in stock market forecasting. The stock market is highly volatile and huge amounts of data are generated every minute globally. The extraction of effective intelligence from this data is of critical importance. However, a collaboration of numerical stock data with qualitative text data can be a challenging task. In this work, we accomplish this and provide an unprecedented, publicly available dataset with technical and fundamental data, sentiment that we gathered from News Archives, TV news captions, Radio Transcripts, Tweets, Daily financial newspapers, etc. The text data entries used for sentiment extraction total more than 1.4 Million. The dataset consists of daily entries from January 2018 to December 2022 for 8 companies representing diverse industrial sectors and the Dow Jones Industrial Average (DJIA) as a whole. Holistic Fundamental and Technical data is provided training ready for Model learning and deployment. The data generated could be used for Incremental online learning with real-time data points retrieved daily, since there was no stagnant data utilized, all the data was retired from APIs or self-designed scripts. Moreover, the utilization of Spearman's rank correlation over real-time data, linking stock returns with sentiment analysis has produced noteworthy results for the DJIA achieving accuracy levels surpassing 60\%. The dataset is made available at https://github.com/batking24/Huge-Stock-Dataset
The popularity of transformer-based text embeddings calls for better statistical tools for measuring distributions of such embeddings. One such tool would be a method for ranking texts within a corpus by centrality, i.e. assigning each text a number signifying how representative that text is of the corpus as a whole. However, an intrinsic center-outward ordering of high-dimensional text representations is not trivial. A statistical depth is a function for ranking k-dimensional objects by measuring centrality with respect to some observed k-dimensional distribution. We adopt a statistical depth to measure distributions of transformer-based text embeddings, transformer-based text embedding (TTE) depth, and introduce the practical use of this depth for both modeling and distributional inference in NLP pipelines. We first define TTE depth and an associated rank sum test for determining whether two corpora differ significantly in embedding space. We then use TTE depth for the task of in-context learning prompt selection, showing that this approach reliably improves performance over statistical baseline approaches across six text classification tasks. Finally, we use TTE depth and the associated rank sum test to characterize the distributions of synthesized and human-generated corpora, showing that five recent synthetic data augmentation processes cause a measurable distributional shift away from associated human-generated text.
Retrieving textual information from natural scene images is an active research area in the field of computer vision with numerous practical applications. Detecting text regions and extracting text from signboards is a challenging problem due to special characteristics like reflecting lights, uneven illumination, or shadows found in real-life natural scene images. With the advent of deep learning-based methods, different sophisticated techniques have been proposed for text detection and text recognition from the natural scene. Though a significant amount of effort has been devoted to extracting natural scene text for resourceful languages like English, little has been done for low-resource languages like Bangla. In this research work, we have proposed an end-to-end system with deep learning-based models for efficiently detecting, recognizing, correcting, and parsing address information from Bangla signboards. We have created manually annotated datasets and synthetic datasets to train signboard detection, address text detection, address text recognition, address text correction, and address text parser models. We have conducted a comparative study among different CTC-based and Encoder-Decoder model architectures for Bangla address text recognition. Moreover, we have designed a novel address text correction model using a sequence-to-sequence transformer-based network to improve the performance of Bangla address text recognition model by post-correction. Finally, we have developed a Bangla address text parser using the state-of-the-art transformer-based pre-trained language model.
Neural knowledge-to-text generation models often struggle to faithfully generate descriptions for the input facts: they may produce hallucinations that contradict the given facts, or describe facts not present in the input. To reduce hallucinations, we propose a novel decoding method, TWEAK (Think While Effectively Articulating Knowledge). TWEAK treats the generated sequences at each decoding step and its future sequences as hypotheses, and ranks each generation candidate based on how well their corresponding hypotheses support the input facts using a Hypothesis Verification Model (HVM). We first demonstrate the effectiveness of TWEAK by using a Natural Language Inference (NLI) model as the HVM and report improved faithfulness with minimal impact on the quality. We then replace the NLI model with our task-specific HVM trained with a first-of-a-kind dataset, FATE (Fact-Aligned Textual Entailment), which pairs input facts with their faithful and hallucinated descriptions with the hallucinated spans marked. The new HVM improves the faithfulness and the quality further and runs faster. Overall the best TWEAK variants improve on average 2.22/7.17 points on faithfulness measured by FactKB over WebNLG and TekGen/GenWiki, respectively, with only 0.14/0.32 points degradation on quality measured by BERTScore over the same datasets. Since TWEAK is a decoding-only approach, it can be integrated with any neural generative model without retraining.
Large language models (LLMs) such as ChatGPT have received immense interest for their general-purpose language understanding and, in particular, their ability to generate high-quality text or computer code. For many professions, LLMs represent an invaluable tool that can speed up and improve the quality of work. In this note, we discuss to what extent they can aid professional mathematicians. We first provide a mathematical description of the transformer model used in all modern language models. Based on recent studies, we then outline best practices and potential issues and report on the mathematical abilities of language models. Finally, we shed light on the potential of LMMs to change how mathematicians work.
The main task of Multimodal Emotion Recognition in Conversations (MERC) is to identify the emotions in modalities, e.g., text, audio, image and video, which is a significant development direction for realizing machine intelligence. However, many data in MERC naturally exhibit an imbalanced distribution of emotion categories, and researchers ignore the negative impact of imbalanced data on emotion recognition. To tackle this problem, we systematically analyze it from three aspects: data augmentation, loss sensitivity, and sampling strategy, and propose the Class Boundary Enhanced Representation Learning (CBERL) model. Concretely, we first design a multimodal generative adversarial network to address the imbalanced distribution of {emotion} categories in raw data. Secondly, a deep joint variational autoencoder is proposed to fuse complementary semantic information across modalities and obtain discriminative feature representations. Finally, we implement a multi-task graph neural network with mask reconstruction and classification optimization to solve the problem of overfitting and underfitting in class boundary learning, and achieve cross-modal emotion recognition. We have conducted extensive experiments on the IEMOCAP and MELD benchmark datasets, and the results show that CBERL has achieved a certain performance improvement in the effectiveness of emotion recognition. Especially on the minority class fear and disgust emotion labels, our model improves the accuracy and F1 value by 10% to 20%.
Through additional training, we explore embedding specialized scientific knowledge into the Llama 2 Large Language Model (LLM). Key findings reveal that effective knowledge integration requires reading texts from multiple perspectives, especially in instructional formats. We utilize text augmentation to tackle the scarcity of specialized texts, including style conversions and translations. Hyperparameter optimization proves crucial, with different size models (7b, 13b, and 70b) reasonably undergoing additional training. Validating our methods, we construct a dataset of 65,000 scientific papers. Although we have succeeded in partially embedding knowledge, the study highlights the complexities and limitations of incorporating specialized information into LLMs, suggesting areas for further improvement.
Generative models for 3D object synthesis have seen significant advancements with the incorporation of prior knowledge distilled from 2D diffusion models. Nevertheless, challenges persist in the form of multi-view geometric inconsistencies and slow generation speeds within the existing 3D synthesis frameworks. This can be attributed to two factors: firstly, the deficiency of abundant geometric a priori knowledge in optimization, and secondly, the entanglement issue between geometry and texture in conventional 3D generation methods.In response, we introduce MetaDreammer, a two-stage optimization approach that leverages rich 2D and 3D prior knowledge. In the first stage, our emphasis is on optimizing the geometric representation to ensure multi-view consistency and accuracy of 3D objects. In the second stage, we concentrate on fine-tuning the geometry and optimizing the texture, thereby achieving a more refined 3D object. Through leveraging 2D and 3D prior knowledge in two stages, respectively, we effectively mitigate the interdependence between geometry and texture. MetaDreamer establishes clear optimization objectives for each stage, resulting in significant time savings in the 3D generation process. Ultimately, MetaDreamer can generate high-quality 3D objects based on textual prompts within 20 minutes, and to the best of our knowledge, it is the most efficient text-to-3D generation method. Furthermore, we introduce image control into the process, enhancing the controllability of 3D generation. Extensive empirical evidence confirms that our method is not only highly efficient but also achieves a quality level that is at the forefront of current state-of-the-art 3D generation techniques.