ObitoNet employs a Cross Attention mechanism to integrate multimodal inputs, where Vision Transformers (ViT) extract semantic features from images and a point cloud tokenizer processes geometric information using Farthest Point Sampling (FPS) and K Nearest Neighbors (KNN) for spatial structure capture. The learned multimodal features are fed into a transformer-based decoder for high-resolution point cloud reconstruction. This approach leverages the complementary strengths of both modalities rich image features and precise geometric details ensuring robust point cloud generation even in challenging conditions such as sparse or noisy data.
Graph neural networks are deep neural networks designed for graphs with attributes attached to nodes or edges. The number of research papers in the literature concerning these models is growing rapidly due to their impressive performance on a broad range of tasks. This survey introduces graph neural networks through the encoder-decoder framework and provides examples of decoders for a range of graph analytic tasks. It uses theory and numerous experiments on homogeneous graphs to illustrate the behavior of graph neural networks for different training sizes and degrees of graph complexity.
Phenotypic drug discovery has attracted widespread attention because of its potential to identify bioactive molecules. Transcriptomic profiling provides a comprehensive reflection of phenotypic changes in cellular responses to external perturbations. In this paper, we propose XTransferCDR, a novel generative framework designed for feature decoupling and transferable representation learning across domains. Given a pair of perturbed expression profiles, our approach decouples the perturbation representations from basal states through domain separation encoders and then cross-transfers them in the latent space. The transferred representations are then used to reconstruct the corresponding perturbed expression profiles via a shared decoder. This cross-transfer constraint effectively promotes the learning of transferable drug perturbation representations. We conducted extensive evaluations of our model on multiple datasets, including single-cell transcriptional responses to drugs and single- and combinatorial genetic perturbations. The experimental results show that XTransferCDR achieved better performance than current state-of-the-art methods, showcasing its potential to advance phenotypic drug discovery.
Large Language Models (LLMs) show promising learning and reasoning abilities. Compared to other NLP tasks, multilingual and multi-label emotion evaluation tasks are under-explored in LLMs. In this paper, we present EthioEmo, a multi-label emotion classification dataset for four Ethiopian languages, namely, Amharic (amh), Afan Oromo (orm), Somali (som), and Tigrinya (tir). We perform extensive experiments with an additional English multi-label emotion dataset from SemEval 2018 Task 1. Our evaluation includes encoder-only, encoder-decoder, and decoder-only language models. We compare zero and few-shot approaches of LLMs to fine-tuning smaller language models. The results show that accurate multi-label emotion classification is still insufficient even for high-resource languages such as English, and there is a large gap between the performance of high-resource and low-resource languages. The results also show varying performance levels depending on the language and model type. EthioEmo is available publicly to further improve the understanding of emotions in language models and how people convey emotions through various languages.
U-Net models with encoder, decoder, and skip-connections components have demonstrated effectiveness in a variety of vision tasks. The skip-connections transmit fine-grained information from the encoder to the decoder. It is necessary to maintain the feature maps used by the skip-connections in memory before the decoding stage. Therefore, they are not friendly to devices with limited resource. In this paper, we propose a universal method and architecture to reduce the memory consumption and meanwhile generate enhanced feature maps to improve network performance. To this end, we design a simple but effective Multi-Scale Information Aggregation Module (MSIAM) in the encoder and an Information Enhancement Module (IEM) in the decoder. The MSIAM aggregates multi-scale feature maps into single-scale with less memory. After that, the aggregated feature maps can be expanded and enhanced to multi-scale feature maps by the IEM. By applying the proposed method on NAFNet, a SOTA model in the field of image restoration, we design a memory-efficient and feature-enhanced network architecture, UNet--. The memory demand by the skip-connections in the UNet-- is reduced by 93.3%, while the performance is improved compared to NAFNet. Furthermore, we show that our proposed method can be generalized to multiple visual tasks, with consistent improvements in both memory consumption and network accuracy compared to the existing efficient architectures.
Learned image compression (LIC) methods often employ symmetrical encoder and decoder architectures, evitably increasing decoding time. However, practical scenarios demand an asymmetric design, where the decoder requires low complexity to cater to diverse low-end devices, while the encoder can accommodate higher complexity to improve coding performance. In this paper, we propose an asymmetric lightweight learned image compression (AsymLLIC) architecture with a novel training scheme, enabling the gradual substitution of complex decoding modules with simpler ones. Building upon this approach, we conduct a comprehensive comparison of different decoder network structures to strike a better trade-off between complexity and compression performance. Experiment results validate the efficiency of our proposed method, which not only achieves comparable performance to VVC but also offers a lightweight decoder with only 51.47 GMACs computation and 19.65M parameters. Furthermore, this design methodology can be easily applied to any LIC models, enabling the practical deployment of LIC techniques.
This paper proposes an Incremental Disentanglement-based Environment-Aware zero-shot text-to-speech (TTS) method, dubbed IDEA-TTS, that can synthesize speech for unseen speakers while preserving the acoustic characteristics of a given environment reference speech. IDEA-TTS adopts VITS as the TTS backbone. To effectively disentangle the environment, speaker, and text factors, we propose an incremental disentanglement process, where an environment estimator is designed to first decompose the environmental spectrogram into an environment mask and an enhanced spectrogram. The environment mask is then processed by an environment encoder to extract environment embeddings, while the enhanced spectrogram facilitates the subsequent disentanglement of the speaker and text factors with the condition of the speaker embeddings, which are extracted from the environmental speech using a pretrained environment-robust speaker encoder. Finally, both the speaker and environment embeddings are conditioned into the decoder for environment-aware speech generation. Experimental results demonstrate that IDEA-TTS achieves superior performance in the environment-aware TTS task, excelling in speech quality, speaker similarity, and environmental similarity. Additionally, IDEA-TTS is also capable of the acoustic environment conversion task and achieves state-of-the-art performance.
Spiking neural networks (SNNs) promise energy-efficient data processing by imitating the event-based behavior of biological neurons. In previous work, we introduced the enlarge-likelihood-each-notable-amplitude spiking-neural-network (ELENA-SNN) decoder, a novel decoding algorithm for low-density parity-check (LDPC) codes. The decoder integrates SNNs into belief propagation (BP) decoding by approximating the check node (CN) update equation using SNNs. However, when decoding LDPC codes with a small variable node(VN) degree, the approximation gets too rough, and the ELENA-SNN decoder does not yield good results. This paper introduces the multi-level ELENA-SNN (ML-ELENA-SNN) decoder, which is an extension of the ELENA-SNN decoder. Instead of a single SNN approximating the CN update, multiple SNNs are applied in parallel, resulting in a higher resolution and higher dynamic range of the exchanged messages. We show that the ML-ELENA-SNN decoder performs similarly to the ubiquitous normalized min-sum decoder for the (38400, 30720) regular LDPC code with a VN degree of dv = 3 and a CN degree of dc = 15.
Recent advancements in text-to-3D generation can generate neural radiance fields (NeRFs) with score distillation sampling, enabling 3D asset creation without real-world data capture. With the rapid advancement in NeRF generation quality, protecting the copyright of the generated NeRF has become increasingly important. While prior works can watermark NeRFs in a post-generation way, they suffer from two vulnerabilities. First, a delay lies between NeRF generation and watermarking because the secret message is embedded into the NeRF model post-generation through fine-tuning. Second, generating a non-watermarked NeRF as an intermediate creates a potential vulnerability for theft. To address both issues, we propose Dreamark to embed a secret message by backdooring the NeRF during NeRF generation. In detail, we first pre-train a watermark decoder. Then, the Dreamark generates backdoored NeRFs in a way that the target secret message can be verified by the pre-trained watermark decoder on an arbitrary trigger viewport. We evaluate the generation quality and watermark robustness against image- and model-level attacks. Extensive experiments show that the watermarking process will not degrade the generation quality, and the watermark achieves 90+% accuracy among both image-level attacks (e.g., Gaussian noise) and model-level attacks (e.g., pruning attack).
Achieving joint learning of Salient Object Detection (SOD) and Camouflaged Object Detection (COD) is extremely challenging due to their distinct object characteristics, i.e., saliency and camouflage. The only preliminary research treats them as two contradictory tasks, training models on large-scale labeled data alternately for each task and assessing them independently. However, such task-specific mechanisms fail to meet real-world demands for addressing unknown tasks effectively. To address this issue, in this paper, we pioneer a task-agnostic framework to unify SOD and COD. To this end, inspired by the agreeable nature of binary segmentation for SOD and COD, we propose a Contrastive Distillation Paradigm (CDP) to distil the foreground from the background, facilitating the identification of salient and camouflaged objects amidst their surroundings. To probe into the contribution of our CDP, we design a simple yet effective contextual decoder involving the interval-layer and global context, which achieves an inference speed of 67 fps. Besides the supervised setting, our CDP can be seamlessly integrated into unsupervised settings, eliminating the reliance on extensive human annotations. Experiments on public SOD and COD datasets demonstrate the superiority of our proposed framework in both supervised and unsupervised settings, compared with existing state-of-the-art approaches. Code is available on https://github.com/liuyi1989/Seamless-Detection.
Achieving high-quality shadow removal with strong generalizability is challenging in scenes with complex global illumination. Due to the limited diversity in shadow removal datasets, current methods are prone to overfitting training data, often leading to reduced performance on unseen cases. To address this, we leverage the rich visual priors of a pre-trained Stable Diffusion (SD) model and propose a two-stage fine-tuning pipeline to adapt the SD model for stable and efficient shadow removal. In the first stage, we fix the VAE and fine-tune the denoiser in latent space, which yields substantial shadow removal but may lose some high-frequency details. To resolve this, we introduce a second stage, called the detail injection stage. This stage selectively extracts features from the VAE encoder to modulate the decoder, injecting fine details into the final results. Experimental results show that our method outperforms state-of-the-art shadow removal techniques. The cross-dataset evaluation further demonstrates that our method generalizes effectively to unseen data, enhancing the applicability of shadow removal methods.
With the rapid development of deep learning and computer vision technologies, medical image segmentation plays a crucial role in the early diagnosis of breast cancer. However, due to the characteristics of breast ultrasound images, such as low contrast, speckle noise, and the highly diverse morphology of tumors, existing segmentation methods exhibit significant limitations in terms of accuracy and robustness. To address these challenges, this study proposes a PINN-based and Enhanced Multi-Scale Feature Fusion Network. The network introduces a Hierarchical Aggregation Encoder in the backbone, which efficiently integrates and globally models multi-scale features through several structural innovations and a novel PCAM module. In the decoder section, a Multi-Scale Feature Refinement Decoder is employed, which, combined with a Multi-Scale Supervision Mechanism and a correction module, significantly improves segmentation accuracy and adaptability. Additionally, the loss function incorporating the PINN mechanism introduces physical constraints during the segmentation process, enhancing the model's ability to accurately delineate tumor boundaries. Comprehensive evaluations on two publicly available breast ultrasound datasets, BUSIS and BUSI, demonstrate that the proposed method outperforms previous segmentation approaches in terms of segmentation accuracy and robustness, particularly under conditions of complex noise and low contrast, effectively improving the accuracy and reliability of tumor segmentation. This method provides a more precise and robust solution for computer-aided diagnosis of breast ultrasound images.
Manipulating articulated objects with robotic arms is challenging due to the complex kinematic structure, which requires precise part segmentation for efficient manipulation. In this work, we introduce a novel superpoint-based perception method designed to improve part segmentation in 3D point clouds of articulated objects. We propose a learnable, part-aware superpoint generation technique that efficiently groups points based on their geometric and semantic similarities, resulting in clearer part boundaries. Furthermore, by leveraging the segmentation capabilities of the 2D foundation model SAM, we identify the centers of pixel regions and select corresponding superpoints as candidate query points. Integrating a query-based transformer decoder further enhances our method's ability to achieve precise part segmentation. Experimental results on the GAPartNet dataset show that our method outperforms existing state-of-the-art approaches in cross-category part segmentation, achieving AP50 scores of 77.9% for seen categories (4.4% improvement) and $39.3\%$ for unseen categories (11.6% improvement), with superior results in 5 out of 9 part categories for seen objects and outperforming all previous methods across all part categories for unseen objects.
This paper introduces text2midi, an end-to-end model to generate MIDI files from textual descriptions. Leveraging the growing popularity of multimodal generative approaches, text2midi capitalizes on the extensive availability of textual data and the success of large language models (LLMs). Our end-to-end system harnesses the power of LLMs to generate symbolic music in the form of MIDI files. Specifically, we utilize a pretrained LLM encoder to process captions, which then condition an autoregressive transformer decoder to produce MIDI sequences that accurately reflect the provided descriptions. This intuitive and user-friendly method significantly streamlines the music creation process by allowing users to generate music pieces using text prompts. We conduct comprehensive empirical evaluations, incorporating both automated and human studies, that show our model generates MIDI files of high quality that are indeed controllable by text captions that may include music theory terms such as chords, keys, and tempo. We release the code and music samples on our demo page (https://github.com/AMAAI-Lab/Text2midi) for users to interact with text2midi.
Masked autoencoders (MAE) have recently succeeded in self-supervised vision representation learning. Previous work mainly applied custom-designed (e.g., random, block-wise) masking or teacher (e.g., CLIP)-guided masking and targets. However, they ignore the potential role of the self-training (student) model in giving feedback to the teacher for masking and targets. In this work, we present to integrate Collaborative Masking and Targets for boosting Masked AutoEncoders, namely CMT-MAE. Specifically, CMT-MAE leverages a simple collaborative masking mechanism through linear aggregation across attentions from both teacher and student models. We further propose using the output features from those two models as the collaborative target of the decoder. Our simple and effective framework pre-trained on ImageNet-1K achieves state-of-the-art linear probing and fine-tuning performance. In particular, using ViT-base, we improve the fine-tuning results of the vanilla MAE from 83.6% to 85.7%.
Techniques enabling large language models (LLMs) to "think more" by generating and attending to intermediate reasoning steps have shown promise in solving complex problems. However, the standard approaches generate sequences of discrete tokens immediately before responding, and so they can incur significant latency costs and be challenging to optimize. In this work, we demonstrate that a frozen LLM can be augmented with an offline coprocessor that operates on the model's key-value (kv) cache. This coprocessor augments the cache with a set of latent embeddings designed to improve the fidelity of subsequent decoding. We train this coprocessor using the language modeling loss from the decoder on standard pretraining data, while keeping the decoder itself frozen. This approach enables the model to learn, in an end-to-end differentiable fashion, how to distill additional computation into its kv-cache. Because the decoder remains unchanged, the coprocessor can operate offline and asynchronously, and the language model can function normally if the coprocessor is unavailable or if a given cache is deemed not to require extra computation. We show experimentally that when a cache is augmented, the decoder achieves lower perplexity on numerous subsequent tokens. Furthermore, even without any task-specific training, our experiments demonstrate that cache augmentation consistently reduces perplexity and improves performance across a range of reasoning-intensive tasks.
Recent advancements in video autoencoders (Video AEs) have significantly improved the quality and efficiency of video generation. In this paper, we propose a novel and compact video autoencoder, VidTwin, that decouples video into two distinct latent spaces: Structure latent vectors, which capture overall content and global movement, and Dynamics latent vectors, which represent fine-grained details and rapid movements. Specifically, our approach leverages an Encoder-Decoder backbone, augmented with two submodules for extracting these latent spaces, respectively. The first submodule employs a Q-Former to extract low-frequency motion trends, followed by downsampling blocks to remove redundant content details. The second averages the latent vectors along the spatial dimension to capture rapid motion. Extensive experiments show that VidTwin achieves a high compression rate of 0.20% with high reconstruction quality (PSNR of 28.14 on the MCL-JCV dataset), and performs efficiently and effectively in downstream generative tasks. Moreover, our model demonstrates explainability and scalability, paving the way for future research in video latent representation and generation. Check our project page for more details: https://vidtwin.github.io/.
Medical image segmentation is crucial in assisting medical doctors in making diagnoses and enabling accurate automatic diagnosis. While advanced convolutional neural networks (CNNs) excel in segmenting regions of interest with pixel-level precision, they often struggle with long-range dependencies, which is crucial for enhancing model performance. Conversely, transformer architectures leverage attention mechanisms to excel in handling long-range dependencies. However, the computational complexity of transformers grows quadratically, posing resource-intensive challenges, especially with high-resolution medical images. Recent research aims to combine CNN and transformer architectures to mitigate their drawbacks and enhance performance while keeping resource demands low. Nevertheless, existing approaches have not fully leveraged the strengths of both architectures to achieve high accuracy with low computational requirements. To address this gap, we propose a novel architecture for 2D medical image segmentation (QTSeg) that leverages a feature pyramid network (FPN) as the image encoder, a multi-level feature fusion (MLFF) as the adaptive module between encoder and decoder and a multi-query mask decoder (MQM Decoder) as the mask decoder. In the first step, an FPN model extracts pyramid features from the input image. Next, MLFF is incorporated between the encoder and decoder to adapt features from different encoder stages to the decoder. Finally, an MQM Decoder is employed to improve mask generation by integrating query tokens with pyramid features at all stages of the mask decoder. Our experimental results show that QTSeg outperforms state-of-the-art methods across all metrics with lower computational demands than the baseline and the existing methods. Code is available at https://github.com/tpnam0901/QTSeg (v0.1.0)
Large language models have shown remarkable performance across a wide range of language tasks, owing to their exceptional capabilities in context modeling. The most commonly used method of context modeling is full self-attention, as seen in standard decoder-only Transformers. Although powerful, this method can be inefficient for long sequences and may overlook inherent input structures. To address these problems, an alternative approach is parallel context encoding, which splits the context into sub-pieces and encodes them parallelly. Because parallel patterns are not encountered during training, naively applying parallel encoding leads to performance degradation. However, the underlying reasons and potential mitigations are unclear. In this work, we provide a detailed analysis of this issue and identify that unusually high attention entropy can be a key factor. Furthermore, we adopt two straightforward methods to reduce attention entropy by incorporating attention sinks and selective mechanisms. Experiments on various tasks reveal that these methods effectively lower irregular attention entropy and narrow performance gaps. We hope this study can illuminate ways to enhance context modeling mechanisms.
Code-switching (CS) automatic speech recognition (ASR) faces challenges due to the language confusion resulting from accents, auditory similarity, and seamless language switches. Adaptation on the pre-trained multi-lingual model has shown promising performance for CS-ASR. In this paper, we adapt Whisper, which is a large-scale multilingual pre-trained speech recognition model, to CS from both encoder and decoder parts. First, we propose an encoder refiner to enhance the encoder's capacity of intra-sentence swithching. Second, we propose using two sets of language-aware adapters with different language prompt embeddings to achieve language-specific decoding information in each decoder layer. Then, a fusion module is added to fuse the language-aware decoding. The experimental results using the SEAME dataset show that, compared with the baseline model, the proposed approach achieves a relative MER reduction of 4.1% and 7.2% on the dev_man and dev_sge test sets, respectively, surpassing state-of-the-art methods. Through experiments, we found that the proposed method significantly improves the performance on non-native language in CS speech, indicating that our approach enables Whisper to better distinguish between the two languages.