In recent years, the wood product industry has been facing a skilled labor shortage. The result is more frequent sudden failures, resulting in additional costs for these companies already operating in a very competitive market. Moreover, sawmills are challenging environments for machinery and sensors. Given that experienced machine operators may be able to diagnose defects or malfunctions, one possible way of assisting novice operators is through acoustic monitoring. As a step towards the automation of wood-processing equipment and decision support systems for machine operators, in this paper, we explore using a deep convolutional autoencoder for acoustic anomaly detection of wood planers on a new real-life dataset. Specifically, our convolutional autoencoder with skip connections (Skip-CAE) and our Skip-CAE transformer outperform the DCASE autoencoder baseline, one-class SVM, isolation forest and a published convolutional autoencoder architecture, respectively obtaining an area under the ROC curve of 0.846 and 0.875 on a dataset of real-factory planer sounds. Moreover, we show that adding skip connections and attention mechanism under the form of a transformer encoder-decoder helps to further improve the anomaly detection capabilities.
3D Referring Expression Segmentation (3D-RES) aims to segment point cloud scenes based on a given expression. However, existing 3D-RES approaches face two major challenges: feature ambiguity and intent ambiguity. Feature ambiguity arises from information loss or distortion during point cloud acquisition due to limitations such as lighting and viewpoint. Intent ambiguity refers to the model's equal treatment of all queries during the decoding process, lacking top-down task-specific guidance. In this paper, we introduce an Image enhanced Prompt Decoding Network (IPDN), which leverages multi-view images and task-driven information to enhance the model's reasoning capabilities. To address feature ambiguity, we propose the Multi-view Semantic Embedding (MSE) module, which injects multi-view 2D image information into the 3D scene and compensates for potential spatial information loss. To tackle intent ambiguity, we designed a Prompt-Aware Decoder (PAD) that guides the decoding process by deriving task-driven signals from the interaction between the expression and visual features. Comprehensive experiments demonstrate that IPDN outperforms the state-ofthe-art by 1.9 and 4.2 points in mIoU metrics on the 3D-RES and 3D-GRES tasks, respectively.
ChartQA presents significant challenges due to the complex distribution of chart elements and the implicit patterns embedded within the underlying data. In this chapter, we have developed a joint multimodal scene graph for charts, explicitly representing the relationships between chart elements and their associated patterns. Our proposed multimodal scene graph consists of two components: a visual graph and a textual graph, each designed to capture the structural and semantic information within the chart. To unify representations across these different modalities, we introduce a multimodal graph contrastive learning approach that learns unified representations by maximizing similarity between nodes representing the same object across multimodal graphs. The learned graph representations can be seamlessly incorporated into a transformer decoder as a soft prompt. Additionally, given the growing need for Multimodal Large Language Models (MLLMs) in zero-shot scenarios, we have designed Chain-of-Thought (CoT) prompts for MLLMs to reduce hallucinations. We tested both methods on public benchmarks such as ChartQA, OpenCQA, and ChartX, demonstrating improved performance and validating the effectiveness of our proposed methods.
Salient Object Detection (SOD) aims to identify and segment prominent regions within a scene. Traditional models rely on manually annotated pseudo labels with precise pixel-level accuracy, which is time-consuming. We developed a low-cost, high-precision annotation method by leveraging large foundation models to address the challenges. Specifically, we use a weakly supervised approach to guide large models in generating pseudo-labels through textual prompts. Since large models do not effectively focus on the salient regions of images, we manually annotate a subset of text to fine-tune the model. Based on this approach, which enables precise and rapid generation of pseudo-labels, we introduce a new dataset, BDS-TR. Compared to the previous DUTS-TR dataset, BDS-TR is more prominent in scale and encompasses a wider variety of categories and scenes. This expansion will enhance our model's applicability across a broader range of scenarios and provide a more comprehensive foundational dataset for future SOD research. Additionally, we present an edge decoder based on dynamic upsampling, which focuses on object edges while gradually recovering image feature resolution. Comprehensive experiments on five benchmark datasets demonstrate that our method significantly outperforms state-of-the-art approaches and also surpasses several existing fully-supervised SOD methods. The code and results will be made available.
Recent advancements in omnimodal learning have been achieved in understanding and generation across images, text, and speech, though mainly within proprietary models. Limited omnimodal datasets and the inherent challenges associated with real-time emotional speech generation have hindered open-source progress. To address these issues, we propose openomni, a two-stage training method combining omnimodal alignment and speech generation to develop a state-of-the-art omnimodal large language model. In the alignment phase, a pre-trained speech model is further trained on text-image tasks to generalize from vision to speech in a (near) zero-shot manner, outperforming models trained on tri-modal datasets. In the speech generation phase, a lightweight decoder facilitates real-time emotional speech through training on speech tasks and preference learning. Experiments demonstrate that openomni consistently improves across omnimodal, vision-language, and speech-language evaluations, enabling natural, emotion-rich dialogues and real-time emotional speech generation.
In the realm of fractal geometry, intricate structures emerge from simple iterative processes that partition parameter spaces into regions of stability and instability. Likewise, training large language models involves iteratively applying update functions, such as Adam, where even slight hyperparameter adjustments can shift the training process from convergence to divergence. Recent evidence from miniature neural networks suggests that the boundary separating these outcomes displays fractal characteristics [1]. Building on these insights, this study extends them to medium-sized, decoder-only transformer architectures by employing a more consistent convergence measure and examining the learning rate hyperparameter landscape for attention and fully connected layers. The results show that the trainability frontier is not a simple threshold; rather, it forms a self-similar yet seemingly random structure at multiple scales, with statistically consistent and repeating patterns. Within this landscape, a region of stable convergence is surrounded by a complex chaotic border, illustrating the sensitive nature of the underlying training dynamics.
In this paper, we consider model order reduction (MOR) methods for problems with slowly decaying Kolmogorov $n$-widths as, e.g., certain wave-like or transport-dominated problems. To overcome this Kolmogorov barrier within MOR, nonlinear projections are used, which are often realized numerically using autoencoders. These autoencoders generally consist of a nonlinear encoder and a nonlinear decoder and involve costly training of the hyperparameters to obtain a good approximation quality of the reduced system. To facilitate the training process, we show that extending the to-be-reduced system and its corresponding training data makes it possible to replace the nonlinear encoder with a linear encoder without sacrificing accuracy, thus roughly halving the number of hyperparameters to be trained.
One of the current trends in robotics is to employ large language models (LLMs) to provide non-predefined command execution and natural human-robot interaction. It is useful to have an environment map together with its language representation, which can be further utilized by LLMs. Such a comprehensive scene representation enables numerous ways of interaction with the map for autonomously operating robots. In this work, we present an approach that enhances incremental implicit mapping through the integration of vision-language features. Specifically, we (i) propose a decoder optimization technique for implicit language maps which can be used when new objects appear on the scene, and (ii) address the problem of inconsistent vision-language predictions between different viewing positions. Our experiments demonstrate the effectiveness of LiLMaps and solid improvements in performance.
Following the remarkable success of Large Language Models (LLMs) in NLP tasks, there is increasing interest in extending their capabilities to speech -- the most common form of communication. The most widespread approach to integrating speech into LLMs is dense feature prepending (DFP), which prepends the projected speech representations to the textual representations, allowing end-to-end training with a speech encoder. This raises questions about the need for a sophisticated speech encoder for DFP and how its performance compares with a standard encoder-decoder (i.e., cross-attention) architecture. We compare DFP and cross-attention under a variety of configurations, such as CTC compression, sequence-level knowledge distillation, on monolingual, bilingual, and multilingual models. To perform a controlled architectural comparison, we train all models from scratch rather than using large pretrained models and use comparable data and parameter settings, testing speech-to-text recognition (ASR) and translation (ST) on MuST-C v1.0 and CoVoST2 datasets. Despite the wide adoption of DFP, our results do not indicate a clear advantage of DFP over cross-attention.
Semantic encoders and decoders for digital semantic communication (SC) often struggle to adapt to variations in unpredictable channel environments and diverse system designs. To address these challenges, this paper proposes a novel framework for training semantic encoders and decoders to enable channel-adaptive digital SC. The core idea is to use binary symmetric channel (BSC) as a universal representation of generic digital communications, eliminating the need to specify channel environments or system designs. Based on this idea, our framework employs parallel BSCs to equivalently model the relationship between the encoder's output and the decoder's input. The bit-flip probabilities of these BSCs are treated as trainable parameters during end-to-end training, with varying levels of regularization applied to address diverse requirements in practical systems. The advantage of our framework is justified by developing a training-aware communication strategy for the inference stage. This strategy makes communication bit errors align with the pre-trained bit-flip probabilities by adaptively selecting power and modulation levels based on practical requirements and channel conditions. Simulation results demonstrate that the proposed framework outperforms existing training approaches in terms of both task performance and power consumption.
Vector quantization is a fundamental technique for compression and large-scale nearest neighbor search. For high-accuracy operating points, multi-codebook quantization associates data vectors with one element from each of multiple codebooks. An example is residual quantization (RQ), which iteratively quantizes the residual error of previous steps. Dependencies between the different parts of the code are, however, ignored in RQ, which leads to suboptimal rate-distortion performance. QINCo recently addressed this inefficiency by using a neural network to determine the quantization codebook in RQ based on the vector reconstruction from previous steps. In this paper we introduce QINCo2 which extends and improves QINCo with (i) improved vector encoding using codeword pre-selection and beam-search, (ii) a fast approximate decoder leveraging codeword pairs to establish accurate short-lists for search, and (iii) an optimized training procedure and network architecture. We conduct experiments on four datasets to evaluate QINCo2 for vector compression and billion-scale nearest neighbor search. We obtain outstanding results in both settings, improving the state-of-the-art reconstruction MSE by 34% for 16-byte vector compression on BigANN, and search accuracy by 24% with 8-byte encodings on Deep1M.
We present `Revelio', a real-world screen-camera communication system leveraging temporal flicker fusion in the OKLAB color space. Using spatially-adaptive flickering and encoding information in pixel region shapes, Revelio achieves visually imperceptible data embedding while remaining robust against noise, asynchronicity, and distortions in screen-camera channels, ensuring reliable decoding by standard smartphone cameras. The decoder, driven by a two-stage neural network, uses a weighted differential accumulator for precise frame detection and symbol recognition. Initial experiments demonstrate Revelio's effectiveness in interactive television, offering an unobtrusive method for meta-information transmission.
With the recent advancements in vision-language models (VLMs) driven by large language models (LLMs), many researchers have focused on models that comprised of an image encoder, an image-to-language projection layer, and a text decoder architectures, leading to the emergence of works like LLava-Med. However, these works primarily operate at the whole-image level, aligning general information from 2D medical images without attending to finer details. As a result, these models often provide irrelevant or non-clinically valuable information while missing critical details. Medical vision-language tasks differ significantly from general images, particularly in their focus on fine-grained details, while excluding irrelevant content. General domain VLMs tend to prioritize global information due to their design, which compresses the entire image into a multi-token representation that is passed into the LLM decoder. Therefore, current VLMs all lack the capability to restrict their attention to particular areas. To address this critical issue in the medical domain, we introduce MedVP, an visual prompt generation and fine-tuning framework, which involves extract medical entities, generate visual prompts, and adapt datasets for visual prompt guided fine-tuning. To the best of our knowledge, this is the first work to explicitly introduce visual prompt into medical VLMs, and we successfully outperform recent state-of-the-art large models across multiple medical VQA datasets. Extensive experiments are conducted to analyze the impact of different visual prompt forms and how they contribute to performance improvement. The results demonstrate both the effectiveness and clinical significance of our approach
Automatic Music Transcription (AMT), aiming to get musical notes from raw audio, typically uses frame-level systems with piano-roll outputs or language model (LM)-based systems with note-level predictions. However, frame-level systems require manual thresholding, while the LM-based systems struggle with long sequences. In this paper, we propose a hybrid method combining pre-trained roll-based encoders with an LM decoder to leverage the strengths of both methods. Besides, our approach employs a hierarchical prediction strategy, first predicting onset and pitch, then velocity, and finally offset. The hierarchical prediction strategy reduces computational costs by breaking down long sequences into different hierarchies. Evaluated on two benchmark roll-based encoders, our method outperforms traditional piano-roll outputs 0.01 and 0.022 in onset-offset-velocity F1 score, demonstrating its potential as a performance-enhancing plug-in for arbitrary roll-based music transcription encoder.
Training transformer-based encoder-decoder models for long document summarization poses a significant challenge due to the quadratic memory consumption during training. Several approaches have been proposed to extend the input length at test time, but training with these approaches is still difficult, requiring truncation of input documents and causing a mismatch between training and test conditions. In this work, we propose CachED (Gradient $\textbf{Cach}$ing for $\textbf{E}$ncoder-$\textbf{D}$ecoder models), an approach that enables end-to-end training of existing transformer-based encoder-decoder models, using the entire document without truncation. Specifically, we apply non-overlapping sliding windows to input documents, followed by fusion in decoder. During backpropagation, the gradients are cached at the decoder and are passed through the encoder in chunks by re-computing the hidden vectors, similar to gradient checkpointing. In the experiments on long document summarization, we extend BART to CachED BART, processing more than 500K tokens during training and achieving superior performance without using any additional parameters.
This paper provides a novel 3D medical image segmentation model structure called nnY-Net. This name comes from the fact that our model adds a cross-attention module at the bottom of the U-net structure to form a Y structure. We integrate the advantages of the two latest SOTA models, MedNeXt and SwinUNETR, and use Swin Transformer as the encoder and ConvNeXt as the decoder to innovatively design the Swin-NeXt structure. Our model uses the lowest-level feature map of the encoder as Key and Value and uses patient features such as pathology and treatment information as Query to calculate the attention weights in a Cross Attention module. Moreover, we simplify some pre- and post-processing as well as data enhancement methods in 3D image segmentation based on the dynUnet and nnU-net frameworks. We integrate our proposed Swin-NeXt with Cross-Attention framework into this framework. Last, we construct a DiceFocalCELoss to improve the training efficiency for the uneven data convergence of voxel classification.
Simultaneous generation models write generation results while reading streaming inputs, necessitating a policy-maker to determine the appropriate output timing. Existing simultaneous generation methods generally adopt the traditional encoder-decoder architecture and learn the generation and policy-making capabilities through complex dynamic programming techniques. Although LLMs excel at text generation, they face challenges in taking on the role of policy-makers through traditional training methods, limiting their exploration in simultaneous generation. To overcome these limitations, we propose a novel LLM-driven Simultaneous Generation (LSG) framework, which allows the off-the-shelf LLM to decide the generation timing and produce output concurrently. Specifically, LSG selects the generation policy that minimizes latency as the baseline policy. Referring to the baseline policy, LSG enables the LLM to devise an improved generation policy that better balances latency and generation quality, and writes generation results accordingly. Experiments on simultaneous translation and streaming automatic speech recognition tasks show that our method can achieve state-of-the-art performance utilizing the open-source LLMs and demonstrate practicality in real-world scenarios.
Due to reasons of convenience and lack of tech literacy, transliteration (i.e., Romanizing native scripts instead of using localization tools) is eminently prevalent in the context of low-resource languages such as Sinhala, which have their own writing script. In this study, our focus is on Romanized Sinhala transliteration. We propose two methods to address this problem: Our baseline is a rule-based method, which is then compared against our second method where we approach the transliteration problem as a sequence-to-sequence task akin to the established Neural Machine Translation (NMT) task. For the latter, we propose a Transformer-based Encode-Decoder solution. We witnessed that the Transformer-based method could grab many ad-hoc patterns within the Romanized scripts compared to the rule-based method. The code base associated with this paper is available on GitHub - https://github.com/kasunw22/Sinhala-Transliterator/
A chatbot is an intelligent software application that automates conversations and engages users in natural language through messaging platforms. Leveraging artificial intelligence (AI), chatbots serve various functions, including customer service, information gathering, and casual conversation. Existing virtual assistant chatbots, such as ChatGPT and Gemini, demonstrate the potential of AI in Natural Language Processing (NLP). However, many current solutions rely on predefined APIs, which can result in vendor lock-in and high costs. To address these challenges, this work proposes a chatbot developed using a Sequence-to-Sequence (Seq2Seq) model with an encoder-decoder architecture that incorporates attention mechanisms and Long Short-Term Memory (LSTM) cells. By avoiding predefined APIs, this approach ensures flexibility and cost-effectiveness. The chatbot is trained, validated, and tested on a dataset specifically curated for the tourism sector in Draa-Tafilalet, Morocco. Key evaluation findings indicate that the proposed Seq2Seq model-based chatbot achieved high accuracies: approximately 99.58% in training, 98.03% in validation, and 94.12% in testing. These results demonstrate the chatbot's effectiveness in providing relevant and coherent responses within the tourism domain, highlighting the potential of specialized AI applications to enhance user experience and satisfaction in niche markets.
End-to-end image and video codecs are becoming increasingly competitive, compared to traditional compression techniques that have been developed through decades of manual engineering efforts. These trainable codecs have many advantages over traditional techniques, such as their straightforward adaptation to perceptual distortion metrics and high performance in specific fields thanks to their learning ability. However, current state-of-the-art neural codecs do not fully exploit the benefits of vector quantization and the existence of the entropy gradient in decoding devices. In this paper, we propose to leverage these two properties (vector quantization and entropy gradient) to improve the performance of off-the-shelf codecs. Firstly, we demonstrate that using non-uniform scalar quantization cannot improve performance over uniform quantization. We thus suggest using predefined optimal uniform vector quantization to improve performance. Secondly, we show that the entropy gradient, available at the decoder, is correlated with the reconstruction error gradient, which is not available at the decoder. We therefore use the former as a proxy to enhance compression performance. Our experimental results show that these approaches save between 1 to 3% of the rate for the same quality across various pretrained methods. In addition, the entropy gradient based solution improves traditional codec performance significantly as well.