The rapid rise of real-time communication and large language models has significantly increased the importance of speech compression. Deep learning-based neural speech codecs have outperformed traditional signal-level speech codecs in terms of rate-distortion (RD) performance. Typically, these neural codecs employ an encoder-quantizer-decoder architecture, where audio is first converted into latent code feature representations and then into discrete tokens. However, this architecture exhibits insufficient RD performance due to two main drawbacks: (1) the inadequate performance of the quantizer, challenging training processes, and issues such as codebook collapse; (2) the limited representational capacity of the encoder and decoder, making it difficult to meet feature representation requirements across various bitrates. In this paper, we propose a rate-aware learned speech compression scheme that replaces the quantizer with an advanced channel-wise entropy model to improve RD performance, simplify training, and avoid codebook collapse. We employ multi-scale convolution and linear attention mixture blocks to enhance the representational capacity and flexibility of the encoder and decoder. Experimental results demonstrate that the proposed method achieves state-of-the-art RD performance, obtaining 53.51% BD-Rate bitrate saving in average, and achieves 0.26 BD-VisQol and 0.44 BD-PESQ gains.
Recent approaches in hierarchical text classification (HTC) rely on the capabilities of a pre-trained transformer model and exploit the label semantics and a graph encoder for the label hierarchy. In this paper, we introduce an effective hierarchical text classifier RADAr (Transformer-based Autoregressive Decoder Architecture) that is based only on an off-the-shelf RoBERTa transformer to process the input and a custom autoregressive decoder with two decoder layers for generating the classification output. Thus, unlike existing approaches for HTC, the encoder of RADAr has no explicit encoding of the label hierarchy and the decoder solely relies on the label sequences of the samples observed during training. We demonstrate on three benchmark datasets that RADAr achieves results competitive to the state of the art with less training and inference time. Our model consistently performs better when organizing the label sequences from children to parents versus the inverse, as done in existing HTC approaches. Our experiments show that neither the label semantics nor an explicit graph encoder for the hierarchy is needed. This has strong practical implications for HTC as the architecture has fewer requirements and provides a speed-up by a factor of 2 at inference time. Moreover, training a separate decoder from scratch in conjunction with fine-tuning the encoder allows future researchers and practitioners to exchange the encoder part as new models arise. The source code is available at https://github.com/yousef-younes/RADAr.
The remarkable success of Large Language Models (LLMs) in understanding and generating various data types, such as images and text, has demonstrated their ability to process and extract semantic information across diverse domains. This transformative capability lays the foundation for semantic communications, enabling highly efficient and intelligent communication systems. In this work, we present a novel OFDM-based semantic communication framework for image transmission. We propose an innovative semantic encoder design that leverages the ability of LLMs to extract the meaning of transmitted data rather than focusing on its raw representation. On the receiver side, we design an LLM-based semantic decoder capable of comprehending context and generating the most appropriate representation to fit the given context. We evaluate our proposed system under different scenarios, including Urban Macro-cell environments with varying speed ranges. The evaluation metrics demonstrate that our proposed system reduces the data size 4250 times, while achieving a higher data rate compared to conventional communication methods. This approach offers a robust and scalable solution to unlock the full potential of 6G connectivity.
Modeling real-world spatio-temporal data is exceptionally difficult due to inherent high dimensionality, measurement noise, partial observations, and often expensive data collection procedures. In this paper, we present Sparse Identification of Nonlinear Dynamics with SHallow REcurrent Decoder networks (SINDy-SHRED), a method to jointly solve the sensing and model identification problems with simple implementation, efficient computation, and robust performance. SINDy-SHRED uses Gated Recurrent Units to model the temporal sequence of sparse sensor measurements along with a shallow decoder network to reconstruct the full spatio-temporal field from the latent state space. Our algorithm introduces a SINDy-based regularization for which the latent space progressively converges to a SINDy-class functional, provided the projection remains within the set. In restricting SINDy to a linear model, a Koopman-SHRED model is generated. SINDy-SHRED (i) learns a symbolic and interpretable generative model of a parsimonious and low-dimensional latent space for the complex spatio-temporal dynamics, (ii) discovers new physics models even for well-known physical systems, (iii) achieves provably robust convergence with an observed globally convex loss landscape, and (iv) achieves superior accuracy, data efficiency, and training time, all with fewer model parameters. We conduct systematic experimental studies on PDE data such as turbulent flows, real-world sensor measurements for sea surface temperature, and direct video data. The interpretable SINDy and Koopman models of latent state dynamics enable stable and accurate long-term video predictions, outperforming all current baseline deep learning models in accuracy, training time, and data requirements, including Convolutional LSTM, PredRNN, ResNet, and SimVP.
Multi-modal retrieval-augmented Question Answering (MRAQA), integrating text and images, has gained significant attention in information retrieval (IR) and natural language processing (NLP). Traditional ranking methods rely on small encoder-based language models, which are incompatible with modern decoder-based generative large language models (LLMs) that have advanced various NLP tasks. To bridge this gap, we propose RAMQA, a unified framework combining learning-to-rank methods with generative permutation-enhanced ranking techniques. We first train a pointwise multi-modal ranker using LLaVA as the backbone. Then, we apply instruction tuning to train a LLaMA model for re-ranking the top-k documents using an innovative autoregressive multi-task learning approach. Our generative ranking model generates re-ranked document IDs and specific answers from document candidates in various permutations. Experiments on two MRAQA benchmarks, WebQA and MultiModalQA, show significant improvements over strong baselines, highlighting the effectiveness of our approach. Code and data are available at: https://github.com/TonyBY/RAMQA
Large language models (LLMs) have shown great potential in natural language processing tasks, but their application to machine translation (MT) remains challenging due to pretraining on English-centric data and the complexity of reinforcement learning from human feedback (RLHF). Direct Preference Optimization (DPO) has emerged as a simpler and more efficient alternative, but its performance depends heavily on the quality of preference data. To address this, we propose Confidence-Reward driven Preference Optimization (CRPO), a novel method that combines reward scores with model confidence to improve data selection for fine-tuning. CRPO selects challenging sentence pairs where the model is uncertain or underperforms, leading to more effective learning. While primarily designed for LLMs, CRPO also generalizes to encoder-decoder models like NLLB, demonstrating its versatility. Empirical results show that CRPO outperforms existing methods such as RS-DPO, RSO and MBR score in both translation accuracy and data efficiency.
Crack Segmentation in industrial concrete surfaces is a challenging task because cracks usually exhibit intricate morphology with slender appearances. Traditional segmentation methods often struggle to accurately locate such cracks, leading to inefficiencies in maintenance and repair processes. In this paper, we propose a novel diffusion-based model with a cross-conditional encoder-decoder, named CrossDiff, which is the first to introduce the diffusion probabilistic model for the crack segmentation task. Specifically, CrossDiff integrates a cross-encoder and a cross-decoder into the diffusion model to constitute a cross-shaped diffusion model structure. The cross-encoder enhances the ability to retain crack details and the cross-decoder helps extract the semantic features of cracks. As a result, CrossDiff can better handle slender cracks. Extensive experiments were conducted on five challenging crack datasets including CFD, CrackTree200, DeepCrack, GAPs384, and Rissbilder. The results demonstrate that the proposed CrossDiff model achieves impressive performance, outperforming other state-of-the-art methods by 8.0% in terms of both Dice score and IoU. The code will be open-source soon.
Monocular depth estimation (MDE) is a challenging task in computer vision, often hindered by the cost and scarcity of high-quality labeled datasets. We tackle this challenge using auxiliary datasets from related vision tasks for an alternating training scheme with a shared decoder built on top of a pre-trained vision foundation model, while giving a higher weight to MDE. Through extensive experiments we demonstrate the benefits of incorporating various in-domain auxiliary datasets and tasks to improve MDE quality on average by ~11%. Our experimental analysis shows that auxiliary tasks have different impacts, confirming the importance of task selection, highlighting that quality gains are not achieved by merely adding data. Remarkably, our study reveals that using semantic segmentation datasets as Multi-Label Dense Classification (MLDC) often results in additional quality gains. Lastly, our method significantly improves the data efficiency for the considered MDE datasets, enhancing their quality while reducing their size by at least 80%. This paves the way for using auxiliary data from related tasks to improve MDE quality despite limited availability of high-quality labeled data. Code is available at https://jugit.fz-juelich.de/ias-8/mdeaux.
The notion of rank of a Boolean function has been a cornerstone in the theory of PAC learning, enabling quasipolynomial-time learning algorithms for polynomial-size decision trees. We present a novel characterization of rank, grounded in the well-known Transformer architecture. We show that the rank of a function $f$ corresponds to the minimum number of Chain of Thought (CoT) steps required by a single-layer transformer decoder with hard attention to compute $f$. Based on this characterization we establish tight bounds on the number of CoT steps required for specific problems, showing that $\ell$-fold function composition necessitates exactly $\ell$ CoT steps. Furthermore, we analyze the problem of identifying the position of the $k$-th occurrence of 1 in a Boolean sequence, proving that it requires $k$ CoT steps.
In this paper, we present an encoder-decoder model leveraging Flan-T5 for post-Automatic Speech Recognition (ASR) Generative Speech Error Correction (GenSEC), and we refer to it as FlanEC. We explore its application within the GenSEC framework to enhance ASR outputs by mapping n-best hypotheses into a single output sentence. By utilizing n-best lists from ASR models, we aim to improve the linguistic correctness, accuracy, and grammaticality of final ASR transcriptions. Specifically, we investigate whether scaling the training data and incorporating diverse datasets can lead to significant improvements in post-ASR error correction. We evaluate FlanEC using the HyPoradise dataset, providing a comprehensive analysis of the model's effectiveness in this domain. Furthermore, we assess the proposed approach under different settings to evaluate model scalability and efficiency, offering valuable insights into the potential of instruction-tuned encoder-decoder models for this task.
Delineating farmland boundaries is essential for agricultural management such as crop monitoring and agricultural census. Traditional methods using remote sensing imagery have been efficient but limited in generalisation. The Segment Anything Model (SAM), known for its impressive zero shot performance, has been adapted for remote sensing tasks through prompt learning and fine tuning. Here, we propose a SAM based farmland boundary delineation framework 'fabSAM' that combines a Deeplabv3+ based Prompter and SAM. Also, a fine tuning strategy was introduced to enable SAMs decoder to improve the use of prompt information. Experimental results on the AI4Boundaries and AI4SmallFarms datasets have shown that fabSAM has a significant improvement in farmland region identification and boundary delineation. Compared to zero shot SAM, fabSAM surpassed it by 23.5% and 15.1% in mIOU on the AI4Boundaries and AI4SmallFarms datasets, respectively. For Deeplabv3+, fabSAM outperformed it by 4.9% and 12.5% in mIOU, respectively. These results highlight the effectiveness of fabSAM, which also means that we can more easily obtain the global farmland region and boundary maps from open source satellite image datasets like Sentinel2.
Geostationary Earth Orbit (GEO) satellite communication demonstrates significant advantages in emergency short burst data services. However, unstable satellite networks, particularly those with frequent packet loss, present a severe challenge to accurate image transmission. To address it, we propose a loss-resilient image coding approach that leverages end-to-end optimization in learned image compression (LIC). Our method builds on the channel-wise progressive coding framework, incorporating Spatial-Channel Rearrangement (SCR) on the encoder side and Mask Conditional Aggregation (MCA) on the decoder side to improve reconstruction quality with unpredictable errors. By integrating the Gilbert-Elliot model into the training process, we enhance the model's ability to generalize in real-world network conditions. Extensive evaluations show that our approach outperforms traditional and deep learning-based methods in terms of compression performance and stability under diverse packet loss, offering robust and efficient progressive transmission even in challenging environments. Code is available at https://github.com/NJUVISION/LossResilientLIC.
RGB and thermal image fusion have great potential to exhibit improved semantic segmentation in low-illumination conditions. Existing methods typically employ a two-branch encoder framework for multimodal feature extraction and design complicated feature fusion strategies to achieve feature extraction and fusion for multimodal semantic segmentation. However, these methods require massive parameter updates and computational effort during the feature extraction and fusion. To address this issue, we propose a novel multimodal fusion network (EFNet) based on an early fusion strategy and a simple but effective feature clustering for training efficient RGB-T semantic segmentation. In addition, we also propose a lightweight and efficient multi-scale feature aggregation decoder based on Euclidean distance. We validate the effectiveness of our method on different datasets and outperform previous state-of-the-art methods with lower parameters and computation.
Video Moment Retrieval and Highlight Detection aim to find corresponding content in the video based on a text query. Existing models usually first use contrastive learning methods to align video and text features, then fuse and extract multimodal information, and finally use a Transformer Decoder to decode multimodal information. However, existing methods face several issues: (1) Overlapping semantic information between different samples in the dataset hinders the model's multimodal aligning performance; (2) Existing models are not able to efficiently extract local features of the video; (3) The Transformer Decoder used by the existing model cannot adequately decode multimodal features. To address the above issues, we proposed the LD-DETR model for Video Moment Retrieval and Highlight Detection tasks. Specifically, we first distilled the similarity matrix into the identity matrix to mitigate the impact of overlapping semantic information. Then, we designed a method that enables convolutional layers to extract multimodal local features more efficiently. Finally, we fed the output of the Transformer Decoder back into itself to adequately decode multimodal information. We evaluated LD-DETR on four public benchmarks and conducted extensive experiments to demonstrate the superiority and effectiveness of our approach. Our model outperforms the State-Of-The-Art models on QVHighlight, Charades-STA and TACoS datasets. Our code is available at https://github.com/qingchen239/ld-detr.
Video deblurring is essential task for autonomous driving, facial recognition, and security surveillance. Traditional methods directly estimate motion blur kernels, often introducing artifacts and leading to poor results. Recent approaches utilize the detection of sharp frames within video sequences to enhance deblurring. However, existing datasets rely on fixed number of sharp frames, which may be too restrictive for some applications and may introduce a bias during model training. To address these limitations and enhance domain adaptability, this work first introduces GoPro Random Sharp (GoProRS), a new dataset where the the frequency of sharp frames within the sequence is customizable, allowing more diverse training and testing scenarios. Furthermore, it presents a novel video deblurring model, called SPEINet, that integrates sharp frame features into blurry frame reconstruction through an attention-based encoder-decoder architecture, a lightweight yet robust sharp frame detection and an edge extraction phase. Extensive experimental results demonstrate that SPEINet outperforms state-of-the-art methods across multiple datasets, achieving an average of +3.2% PSNR improvement over recent techniques. Given such promising results, we believe that both the proposed model and dataset pave the way for future advancements in video deblurring based on the detection of sharp frames.
We present ORCAst, a multi-stage, multi-arm network for Operational high-Resolution Current forecAsts over one week. Producing real-time nowcasts and forecasts of ocean surface currents is a challenging problem due to indirect or incomplete information from satellite remote sensing data. Entirely trained on real satellite data and in situ measurements from drifters, our model learns to forecast global ocean surface currents using various sources of ground truth observations in a multi-stage learning procedure. Our multi-arm encoder-decoder model architecture allows us to first predict sea surface height and geostrophic currents from larger quantities of nadir and SWOT altimetry data, before learning to predict ocean surface currents from much more sparse in situ measurements from drifters. Training our model on specific regions improves performance. Our model achieves stronger nowcast and forecast performance in predicting ocean surface currents than various state-of-the-art methods.
Point cloud completion aims to reconstruct complete 3D shapes from partial 3D point clouds. With advancements in deep learning techniques, various methods for point cloud completion have been developed. Despite achieving encouraging results, a significant issue remains: these methods often overlook the variability in point clouds sampled from a single 3D object surface. This variability can lead to ambiguity and hinder the achievement of more precise completion results. Therefore, in this study, we introduce a novel point cloud completion network, namely Dual-Codebook Point Completion Network (DC-PCN), following an encder-decoder pipeline. The primary objective of DC-PCN is to formulate a singular representation of sampled point clouds originating from the same 3D surface. DC-PCN introduces a dual-codebook design to quantize point-cloud representations from a multilevel perspective. It consists of an encoder-codebook and a decoder-codebook, designed to capture distinct point cloud patterns at shallow and deep levels. Additionally, to enhance the information flow between these two codebooks, we devise an information exchange mechanism. This approach ensures that crucial features and patterns from both shallow and deep levels are effectively utilized for completion. Extensive experiments on the PCN, ShapeNet\_Part, and ShapeNet34 datasets demonstrate the state-of-the-art performance of our method.
Visual anomaly detection is a highly challenging task, often categorized as a one-class classification and segmentation problem. Recent studies have demonstrated that the student-teacher (S-T) framework effectively addresses this challenge. However, most S-T frameworks rely solely on pre-trained teacher networks to guide student networks in learning multi-scale similar features, overlooking the potential of the student networks to enhance learning through multi-scale feature fusion. In this study, we propose a novel model named PFADSeg, which integrates a pre-trained teacher network, a denoising student network with multi-scale feature fusion, and a guided anomaly segmentation network into a unified framework. By adopting a unique teacher-encoder and student-decoder denoising mode, the model improves the student network's ability to learn from teacher network features. Furthermore, an adaptive feature fusion mechanism is introduced to train a self-supervised segmentation network that synthesizes anomaly masks autonomously, significantly increasing detection performance. Evaluated on the MVTec AD dataset, PFADSeg achieves state-of-the-art results with an image-level AUC of 98.9%, a pixel-level mean precision of 76.4%, and an instance-level mean precision of 78.7%.
The standard language modeling (LM) loss by itself has been shown to be inadequate for effective dialogue modeling. As a result, various training approaches, such as auxiliary loss functions and leveraging human feedback, are being adopted to enrich open-domain dialogue systems. One such auxiliary loss function is Bag-of-Words (BoW) loss, defined as the cross-entropy loss for predicting all the words/tokens of the next utterance. In this work, we propose a novel auxiliary loss named Bag-of-Keywords (BoK) loss to capture the central thought of the response through keyword prediction and leverage it to enhance the generation of meaningful and interpretable responses in open-domain dialogue systems. BoK loss upgrades the BoW loss by predicting only the keywords or critical words/tokens of the next utterance, intending to estimate the core idea rather than the entire response. We incorporate BoK loss in both encoder-decoder (T5) and decoder-only (DialoGPT) architecture and train the models to minimize the weighted sum of BoK and LM (BoK-LM) loss. We perform our experiments on two popular open-domain dialogue datasets, DailyDialog and Persona-Chat. We show that the inclusion of BoK loss improves the dialogue generation of backbone models while also enabling post-hoc interpretability. We also study the effectiveness of BoK-LM loss as a reference-free metric and observe comparable performance to the state-of-the-art metrics on various dialogue evaluation datasets.
Generative modeling aims to generate new data samples that resemble a given dataset, with diffusion models recently becoming the most popular generative model. One of the main challenges of diffusion models is solving the problem in the input space, which tends to be very high-dimensional. Recently, solving diffusion models in the latent space through an encoder that maps from the data space to a lower-dimensional latent space has been considered to make the training process more efficient and has shown state-of-the-art results. The variational autoencoder (VAE) is the most commonly used encoder/decoder framework in this domain, known for its ability to learn latent representations and generate data samples. In this paper, we introduce a novel encoder/decoder framework with theoretical properties distinct from those of the VAE, specifically designed to preserve the geometric structure of the data distribution. We demonstrate the significant advantages of this geometry-preserving encoder in the training process of both the encoder and decoder. Additionally, we provide theoretical results proving convergence of the training process, including convergence guarantees for encoder training, and results showing faster convergence of decoder training when using the geometry-preserving encoder.