Table structure recognition (TSR) aims to convert tabular images into a machine-readable format, where a visual encoder extracts image features and a textual decoder generates table-representing tokens. Existing approaches use classic convolutional neural network (CNN) backbones for the visual encoder and transformers for the textual decoder. However, this hybrid CNN-Transformer architecture introduces a complex visual encoder that accounts for nearly half of the total model parameters, markedly reduces both training and inference speed, and hinders the potential for self-supervised learning in TSR. In this work, we design a lightweight visual encoder for TSR without sacrificing expressive power. We discover that a convolutional stem can match classic CNN backbone performance, with a much simpler model. The convolutional stem strikes an optimal balance between two crucial factors for high-performance TSR: a higher receptive field (RF) ratio and a longer sequence length. This allows it to "see" an appropriate portion of the table and "store" the complex table structure within sufficient context length for the subsequent transformer. We conducted reproducible ablation studies and open-sourced our code at https://github.com/poloclub/tsr-convstem to enhance transparency, inspire innovations, and facilitate fair comparisons in our domain as tables are a promising modality for representation learning.
Minimum Bayes Risk (MBR) decoding has been shown to be a powerful alternative to beam search decoding in a variety of text generation tasks. MBR decoding selects a hypothesis from a pool of hypotheses that has the least expected risk under a probability model according to a given utility function. Since it is impractical to compute the expected risk exactly over all possible hypotheses, two approximations are commonly used in MBR. First, it integrates over a sampled set of hypotheses rather than over all possible hypotheses. Second, it estimates the probability of each hypothesis using a Monte Carlo estimator. While the first approximation is necessary to make it computationally feasible, the second is not essential since we typically have access to the model probability at inference time. We propose Model-Based MBR (MBMBR), a variant of MBR that uses the model probability itself as the estimate of the probability distribution instead of the Monte Carlo estimate. We show analytically and empirically that the model-based estimate is more promising than the Monte Carlo estimate in text generation tasks. Our experiments show that MBMBR outperforms MBR in several text generation tasks, both with encoder-decoder models and with large language models.
Since its introduction, the transformers architecture has seen great adoption in NLP applications, but it also has limitations. Although the self-attention mechanism allows for generating very rich representations of the input text, its effectiveness may be limited in specialized domains such as legal, where, for example, language models often have to process very long texts. In this paper, we explore alternatives to replace the attention-based layers with simpler token-mixing mechanisms: Hartley and Fourier transforms. Using these non-parametric techniques, we train models with long input documents from scratch in the legal domain setting. We also introduce a new hybrid Seq2Seq architecture, a no-attention-based encoder connected with an attention-based decoder, which performs quite well on existing summarization tasks with much less compute and memory requirements. We believe that similar, if not better performance, as in the case of long correlations of abstractive text summarization tasks, can be achieved by adopting these simpler infrastructures. This not only makes training models from scratch accessible to more people, but also contributes to the reduction of the carbon footprint during training.
This paper explores the integration of deep learning techniques for joint sensing and communications, with an extension to semantic communications. The integrated system comprises a transmitter and receiver operating over a wireless channel, subject to noise and fading. The transmitter employs a deep neural network (DNN), namely an encoder, for joint operations of source coding, channel coding, and modulation, while the receiver utilizes another DNN, namely a decoder, for joint operations of demodulation, channel decoding, and source decoding to reconstruct the data samples. The transmitted signal serves a dual purpose, supporting communication with the receiver and enabling sensing. When a target is present, the reflected signal is received, and another DNN decoder is utilized for sensing. This decoder is responsible for detecting the target's presence and determining its range. All these DNNs, including one encoder and two decoders, undergo joint training through multi-task learning, considering data and channel characteristics. This paper extends to incorporate semantic communications by introducing an additional DNN, another decoder at the receiver, operating as a task classifier. This decoder evaluates the fidelity of label classification for received signals, enhancing the integration of semantics within the communication process. The study presents results based on using the CIFAR-10 as the input data and accounting for channel effects like Additive White Gaussian Noise (AWGN) and Rayleigh fading. The results underscore the effectiveness of multi-task deep learning in achieving high-fidelity joint sensing and semantic communications.
Many recent prompting strategies for large language models (LLMs) query the model multiple times sequentially -- first to produce intermediate results and then the final answer. However, using these methods, both decoder and model are unaware of potential follow-up prompts, leading to disconnected and undesirably wordy intermediate responses. In this work, we address this issue by proposing prompt sketching, a new prompting paradigm in which an LLM does not only respond by completing a prompt, but by predicting values for multiple variables in a template. This way, sketching grants users more control over the generation process, e.g., by providing a reasoning framework via intermediate instructions, leading to better overall results. The key idea enabling sketching with existing, autoregressive models is to adapt the decoding procedure to also score follow-up instructions during text generation, thus optimizing overall template likelihood in inference. Our experiments show that in a zero-shot setting, prompt sketching outperforms existing, sequential prompting schemes such as direct asking or chain-of-thought on 7 out of 8 LLM benchmarking tasks, including state tracking, arithmetic reasoning, and general question answering. To facilitate future use, we release a number of generic, yet effective sketches applicable to many tasks, and an open source library called dclib, powering our sketch-aware decoders.
Recently, unified speech-text models, such as SpeechGPT, VioLA, and AudioPaLM, have achieved remarkable performance on various speech tasks. These models discretize speech signals into tokens (speech discretization) and use a shared vocabulary for both text and speech tokens. Then they train a single decoder-only Transformer on a mixture of speech tasks. However, these models rely on the Loss Masking strategy for the ASR task, which ignores the dependency among speech tokens. In this paper, we propose to model speech tokens in an autoregressive way, similar to text. We find that applying the conventional cross-entropy loss on input speech tokens does not consistently improve the ASR performance over the Loss Masking approach. To address this issue, we propose a novel approach denoted Smoothed Label Distillation (SLD), which applies a KL divergence loss with smoothed labels on speech tokens. Our experiments show that SLD effectively models speech tokens and outperforms Loss Masking for decoder-only Transformers in ASR tasks with different speech discretization methods. The source code can be found here: https://github.com/alibaba-damo-academy/SpokenNLP/tree/main/sld
In this research we propose a deep learning approach for detecting anomalies in videos using convolutional autoencoder and decoder neural networks on the UCSD dataset.Our method utilizes a convolutional autoencoder to learn the spatiotemporal patterns of normal videos and then compares each frame of a test video to this learned representation. We evaluated our approach on the UCSD dataset and achieved an overall accuracy of 99.35% on the Ped1 dataset and 99.77% on the Ped2 dataset, demonstrating the effectiveness of our method for detecting anomalies in surveillance videos. The results show that our method outperforms other state-of-the-art methods, and it can be used in real-world applications for video anomaly detection.
Developing a new Salient Object Detection (SOD) model involves selecting an ImageNet pre-trained backbone and creating novel feature refinement modules to use backbone features. However, adding new components to a pre-trained backbone needs retraining the whole network on the ImageNet dataset, which requires significant time. Hence, we explore developing a neural network from scratch directly trained on SOD without ImageNet pre-training. Such a formulation offers full autonomy to design task-specific components. To that end, we propose SODAWideNet, an encoder-decoder-style network for Salient Object Detection. We deviate from the commonly practiced paradigm of narrow and deep convolutional models to a wide and shallow architecture, resulting in a parameter-efficient deep neural network. To achieve a shallower network, we increase the receptive field from the beginning of the network using a combination of dilated convolutions and self-attention. Therefore, we propose Multi Receptive Field Feature Aggregation Module (MRFFAM) that efficiently obtains discriminative features from farther regions at higher resolutions using dilated convolutions. Next, we propose Multi-Scale Attention (MSA), which creates a feature pyramid and efficiently computes attention across multiple resolutions to extract global features from larger feature maps. Finally, we propose two variants, SODAWideNet-S (3.03M) and SODAWideNet (9.03M), that achieve competitive performance against state-of-the-art models on five datasets.
In large-scale storehouses, precise instance masks are crucial for robotic bin picking but are challenging to obtain. Existing instance segmentation methods typically rely on a tedious process of scene collection, mask annotation, and network fine-tuning for every single Stock Keeping Unit (SKU). This paper presents SKU-Patch, a new patch-guided instance segmentation solution, leveraging only a few image patches for each incoming new SKU to predict accurate and robust masks, without tedious manual effort and model re-training. Technical-wise, we design a novel transformer-based network with (i) a patch-image correlation encoder to capture multi-level image features calibrated by patch information and (ii) a patch-aware transformer decoder with parallel task heads to generate instance masks. Extensive experiments on four storehouse benchmarks manifest that SKU-Patch is able to achieve the best performance over the state-of-the-art methods. Also, SKU-Patch yields an average of nearly 100% grasping success rate on more than 50 unseen SKUs in a robot-aided auto-store logistic pipeline, showing its effectiveness and practicality.
Multi-modal Large Language Models (MLLMs) have demonstrated impressive instruction abilities across various open-ended tasks. However, previous methods primarily focus on enhancing multi-modal capabilities. In this work, we introduce a versatile multi-modal large language model, mPLUG-Owl2, which effectively leverages modality collaboration to improve performance in both text and multi-modal tasks. mPLUG-Owl2 utilizes a modularized network design, with the language decoder acting as a universal interface for managing different modalities. Specifically, mPLUG-Owl2 incorporates shared functional modules to facilitate modality collaboration and introduces a modality-adaptive module that preserves modality-specific features. Extensive experiments reveal that mPLUG-Owl2 is capable of generalizing both text tasks and multi-modal tasks and achieving state-of-the-art performances with a single generic model. Notably, mPLUG-Owl2 is the first MLLM model that demonstrates the modality collaboration phenomenon in both pure-text and multi-modal scenarios, setting a pioneering path in the development of future multi-modal foundation models.
The International Phonetic Alphabet (IPA) is indispensable in language learning and understanding, aiding users in accurate pronunciation and comprehension. Additionally, it plays a pivotal role in speech therapy, linguistic research, accurate transliteration, and the development of text-to-speech systems, making it an essential tool across diverse fields. Bangla being 7th as one of the widely used languages, gives rise to the need for IPA in its domain. Its IPA mapping is too diverse to be captured manually giving the need for Artificial Intelligence and Machine Learning in this field. In this study, we have utilized a transformer-based sequence-to-sequence model at the letter and symbol level to get the IPA of each Bangla word as the variation of IPA in association of different words is almost null. Our transformer model only consisted of 8.5 million parameters with only a single decoder and encoder layer. Additionally, to handle the punctuation marks and the occurrence of foreign languages in the text, we have utilized manual mapping as the model won't be able to learn to separate them from Bangla words while decreasing our required computational resources. Finally, maintaining the relative position of the sentence component IPAs and generation of the combined IPA has led us to achieve the top position with a word error rate of 0.10582 in the public ranking of DataVerse Challenge - ITVerse 2023 (https://www.kaggle.com/competitions/dataverse_2023/).
Automatic prostate segmentation in TRUS images has always been a challenging problem, since prostates in TRUS images have ambiguous boundaries and inhomogeneous intensity distribution. Although many prostate segmentation methods have been proposed, they still need to be improved due to the lack of sensibility to edge information. Consequently, the objective of this study is to devise a highly effective prostate segmentation method that overcomes these limitations and achieves accurate segmentation of prostates in TRUS images. A 3D edge-aware attention generative adversarial network (3D EAGAN)-based prostate segmentation method is proposed in this paper, which consists of an edge-aware segmentation network (EASNet) that performs the prostate segmentation and a discriminator network that distinguishes predicted prostates from real prostates. The proposed EASNet is composed of an encoder-decoder-based U-Net backbone network, a detail compensation module, four 3D spatial and channel attention modules, an edge enhance module, and a global feature extractor. The detail compensation module is proposed to compensate for the loss of detailed information caused by the down-sampling process of the encoder. The features of the detail compensation module are selectively enhanced by the 3D spatial and channel attention module. Furthermore, an edge enhance module is proposed to guide shallow layers in the EASNet to focus on contour and edge information in prostates. Finally, features from shallow layers and hierarchical features from the decoder module are fused through the global feature extractor to predict the segmentation prostates.
Deepfake videos, generated through AI faceswapping techniques, have garnered considerable attention due to their potential for powerful impersonation attacks. While existing research primarily focuses on binary classification to discern between real and fake videos, however determining the specific generation model for a fake video is crucial for forensic investigation. Addressing this gap, this paper investigates the model attribution problem of Deepfake videos from a recently proposed dataset, Deepfakes from Different Models (DFDM), derived from various Autoencoder models. The dataset comprises 6,450 Deepfake videos generated by five distinct models with variations in encoder, decoder, intermediate layer, input resolution, and compression ratio. This study formulates Deepfakes model attribution as a multiclass classification task, proposing a segment of VGG19 as a feature extraction backbone, known for its effectiveness in imagerelated tasks, while integrated a Capsule Network with a Spatio-Temporal attention mechanism. The Capsule module captures intricate hierarchies among features for robust identification of deepfake attributes. Additionally, the video-level fusion technique leverages temporal attention mechanisms to handle concatenated feature vectors, capitalizing on inherent temporal dependencies in deepfake videos. By aggregating insights across frames, our model gains a comprehensive understanding of video content, resulting in more precise predictions. Experimental results on the deepfake benchmark dataset (DFDM) demonstrate the efficacy of our proposed method, achieving up to a 4% improvement in accurately categorizing deepfake videos compared to baseline models while demanding fewer computational resources.
We present a novel usage of Transformers to make image classification interpretable. Unlike mainstream classifiers that wait until the last fully connected layer to incorporate class information to make predictions, we investigate a proactive approach, asking each class to search for itself in an image. We realize this idea via a Transformer encoder-decoder inspired by DEtection TRansformer (DETR). We learn "class-specific" queries (one for each class) as input to the decoder, enabling each class to localize its patterns in an image via cross-attention. We name our approach INterpretable TRansformer (INTR), which is fairly easy to implement and exhibits several compelling properties. We show that INTR intrinsically encourages each class to attend distinctively; the cross-attention weights thus provide a faithful interpretation of the prediction. Interestingly, via "multi-head" cross-attention, INTR could identify different "attributes" of a class, making it particularly suitable for fine-grained classification and analysis, which we demonstrate on eight datasets. Our code and pre-trained models are publicly accessible at the Imageomics Institute GitHub site: https://github.com/Imageomics/INTR.
Interactive and embodied tasks pose at least two fundamental challenges to existing Vision & Language (VL) models, including 1) grounding language in trajectories of actions and observations, and 2) referential disambiguation. To tackle these challenges, we propose an Embodied MultiModal Agent (EMMA): a unified encoder-decoder model that reasons over images and trajectories, and casts action prediction as multimodal text generation. By unifying all tasks as text generation, EMMA learns a language of actions which facilitates transfer across tasks. Different to previous modular approaches with independently trained components, we use a single multitask model where each task contributes to goal completion. EMMA performs on par with similar models on several VL benchmarks and sets a new state-of-the-art performance (36.81% success rate) on the Dialog-guided Task Completion (DTC), a benchmark to evaluate dialog-guided agents in the Alexa Arena
Self-supervised learning has been actively studied in time series domain recently, especially for masked reconstruction. Most of these methods follow the "Pre-training + Fine-tuning" paradigm in which a new decoder replaces the pre-trained decoder to fit for a specific downstream task, leading to inconsistency of upstream and downstream tasks. In this paper, we first point out that the unification of task objectives and adaptation for task difficulty are critical for bridging the gap between time series masked reconstruction and forecasting. By reserving the pre-trained mask token during fine-tuning stage, the forecasting task can be taken as a special case of masked reconstruction, where the future values are masked and reconstructed based on history values. It guarantees the consistency of task objectives but there is still a gap in task difficulty. Because masked reconstruction can utilize contextual information while forecasting can only use historical information to reconstruct. To further mitigate the existed gap, we propose a simple yet effective prompt token tuning (PT-Tuning) paradigm, in which all pre-trained parameters are frozen and only a few trainable prompt tokens are added to extended mask tokens in element-wise manner. Extensive experiments on real-world datasets demonstrate the superiority of our proposed paradigm with state-of-the-art performance compared to representation learning and end-to-end supervised forecasting methods.
Referring image segmentation aims to segment the target object described by a natural-language utterance. Recent approaches typically distinguish pixels by aligning pixel-wise visual features with linguistic features extracted from the referring description. Nevertheless, such a free-form description only specifies certain discriminative attributes of the target object or its relations to a limited number of objects, which fails to represent the rich visual context adequately. The stand-alone linguistic features are therefore unable to align with all visual concepts, resulting in inaccurate segmentation. In this paper, we propose to address this issue by incorporating rich visual context into linguistic features for sufficient vision-language alignment. Specifically, we present Context-Aware Referring Image Segmentation (CARIS), a novel architecture that enhances the contextual awareness of linguistic features via sequential vision-language attention and learnable prompts. Technically, CARIS develops a context-aware mask decoder with sequential bidirectional cross-modal attention to integrate the linguistic features with visual context, which are then aligned with pixel-wise visual features. Furthermore, two groups of learnable prompts are employed to delve into additional contextual information from the input image and facilitate the alignment with non-target pixels, respectively. Extensive experiments demonstrate that CARIS achieves new state-of-the-art performances on three public benchmarks. Code is available at https://github.com/lsa1997/CARIS.
Contemporary translation engines based on the encoder-decoder framework have made significant strides in development. However, the emergence of Large Language Models (LLMs) has disrupted their position by presenting the potential for achieving superior translation quality. To uncover the circumstances in which LLMs excel and explore how their strengths can be harnessed to enhance translation quality, we first conduct a comprehensive analysis to assess the strengths and limitations of various commercial NMT systems and MT-oriented LLMs. Our findings indicate that neither NMT nor MT-oriented LLMs alone can effectively address all the translation issues, but MT-oriented LLMs show promise as a complementary solution to NMT systems. Building upon these insights, we propose Cooperative Decoding (CoDec), which treats NMT systems as a pretranslation model and MT-oriented LLMs as a supplemental solution to handle complex scenarios beyond the capability of NMT alone. Experimental results on the WMT22 test sets and a newly collected test set WebCrawl demonstrate the effectiveness and efficiency of CoDec, highlighting its potential as a robust solution for combining NMT systems with MT-oriented LLMs in the field of machine translation.
This paper presents the Ensemble Nucleotide Byte-level Encoder-Decoder (ENBED) foundation model, analyzing DNA sequences at byte-level precision with an encoder-decoder Transformer architecture. ENBED uses a sub-quadratic implementation of attention to develop an efficient model capable of sequence-to-sequence transformations, generalizing previous genomic models with encoder-only or decoder-only architectures. We use Masked Language Modeling to pre-train the foundation model using reference genome sequences and apply it in the following downstream tasks: (1) identification of enhancers, promotors and splice sites, (2) recognition of sequences containing base call mismatches and insertion/deletion errors, an advantage over tokenization schemes involving multiple base pairs, which lose the ability to analyze with byte-level precision, (3) identification of biological function annotations of genomic sequences, and (4) generating mutations of the Influenza virus using the encoder-decoder architecture and validating them against real-world observations. In each of these tasks, we demonstrate significant improvement as compared to the existing state-of-the-art results.
Manga, a widely celebrated Japanese comic art form, is renowned for its diverse narratives and distinct artistic styles. However, the inherently visual and intricate structure of Manga, which comprises images housing multiple panels, poses significant challenges for content retrieval. To address this, we present MaRU (Manga Retrieval and Understanding), a multi-staged system that connects vision and language to facilitate efficient search of both dialogues and scenes within Manga frames. The architecture of MaRU integrates an object detection model for identifying text and frame bounding boxes, a Vision Encoder-Decoder model for text recognition, a text encoder for embedding text, and a vision-text encoder that merges textual and visual information into a unified embedding space for scene retrieval. Rigorous evaluations reveal that MaRU excels in end-to-end dialogue retrieval and exhibits promising results for scene retrieval.