In this paper, we construct two research objectives: i) explore the learned embedding space of BiomedCLIP, an open-source large vision language model, to analyse meaningful class separations, and ii) quantify the limitations of BiomedCLIP when applied to a highly imbalanced, out-of-distribution multi-label medical dataset. We experiment on IU-xray dataset, which exhibits the aforementioned criteria, and evaluate BiomedCLIP in classifying images (radiographs) in three contexts: zero-shot inference, full finetuning, and linear probing. The results show that the model under zero-shot settings over-predicts all labels, leading to poor precision and inter-class separability. Full fine-tuning improves classification of distinct diseases, while linear probing detects overlapping features. We demonstrate visual understanding of the model using Grad-CAM heatmaps and compare with 15 annotations by a radiologist. We highlight the need for careful adaptations of the models to foster reliability and applicability in a real-world setting. The code for the experiments in this work is available and maintained on GitHub.
Document Visual Question Answering (DocVQA) is a practical yet challenging task, which is to ask questions based on documents while referring to multiple pages and different modalities of information, e.g, images and tables. To handle multi-modality, recent methods follow a similar Retrieval Augmented Generation (RAG) pipeline, but utilize Visual Language Models (VLMs) based embedding model to embed and retrieve relevant pages as images, and generate answers with VLMs that can accept an image as input. In this paper, we introduce SimpleDoc, a lightweight yet powerful retrieval - augmented framework for DocVQA. It boosts evidence page gathering by first retrieving candidates through embedding similarity and then filtering and re-ranking these candidates based on page summaries. A single VLM-based reasoner agent repeatedly invokes this dual-cue retriever, iteratively pulling fresh pages into a working memory until the question is confidently answered. SimpleDoc outperforms previous baselines by 3.2% on average on 4 DocVQA datasets with much fewer pages retrieved. Our code is available at https://github.com/ag2ai/SimpleDoc.
Graph-based Retrieval-Augmented Generation (RAG) has shown great capability in enhancing Large Language Model (LLM)'s answer with an external knowledge base. Compared to traditional RAG, it introduces a graph as an intermediate representation to capture better structured relational knowledge in the corpus, elevating the precision and comprehensiveness of generation results. However, developers usually face challenges in analyzing the effectiveness of GraphRAG on their dataset due to GraphRAG's complex information processing pipeline and the overwhelming amount of LLM invocations involved during graph construction and query, which limits GraphRAG interpretability and accessibility. This research proposes a visual analysis framework that helps RAG developers identify critical recalls of GraphRAG and trace these recalls through the GraphRAG pipeline. Based on this framework, we develop XGraphRAG, a prototype system incorporating a set of interactive visualizations to facilitate users' analysis process, boosting failure cases collection and improvement opportunities identification. Our evaluation demonstrates the effectiveness and usability of our approach. Our work is open-sourced and available at https://github.com/Gk0Wk/XGraphRAG.
The emergence of GPT-4o-like large multimodal models (LMMs) has raised the exploration of integrating text, vision, and speech modalities to support more flexible multimodal interaction. Existing LMMs typically concatenate representation of modalities along the sequence dimension and feed them into a large language model (LLM) backbone. While sequence-dimension concatenation is straightforward for modality integration, it often relies heavily on large-scale data to learn modality alignments. In this paper, we aim to model the relationships between modalities more purposefully, thereby achieving more efficient and flexible modality alignments. To this end, we propose Stream-Omni, a large language-vision-speech model with efficient modality alignments, which can simultaneously support interactions under various modality combinations. Stream-Omni employs LLM as the backbone and aligns the vision and speech to the text based on their relationships. For vision that is semantically complementary to text, Stream-Omni uses sequence-dimension concatenation to achieve vision-text alignment. For speech that is semantically consistent with text, Stream-Omni introduces a CTC-based layer-dimension mapping to achieve speech-text alignment. In this way, Stream-Omni can achieve modality alignments with less data (especially speech), enabling the transfer of text capabilities to other modalities. Experiments on various benchmarks demonstrate that Stream-Omni achieves strong performance on visual understanding, speech interaction, and vision-grounded speech interaction tasks. Owing to the layer-dimensional mapping, Stream-Omni can simultaneously provide intermediate text outputs (such as ASR transcriptions and model responses) during speech interaction, offering users a comprehensive multimodal experience.
Existing large-scale zero-shot text-to-speech (TTS) models deliver high speech quality but suffer from slow inference speeds due to massive parameters. To address this issue, this paper introduces ZipVoice, a high-quality flow-matching-based zero-shot TTS model with a compact model size and fast inference speed. Key designs include: 1) a Zipformer-based flow-matching decoder to maintain adequate modeling capabilities under constrained size; 2) Average upsampling-based initial speech-text alignment and Zipformer-based text encoder to improve speech intelligibility; 3) A flow distillation method to reduce sampling steps and eliminate the inference overhead associated with classifier-free guidance. Experiments on 100k hours multilingual datasets show that ZipVoice matches state-of-the-art models in speech quality, while being 3 times smaller and up to 30 times faster than a DiT-based flow-matching baseline. Codes, model checkpoints and demo samples are publicly available.
As an emerging paradigm of brain-computer interfaces (BCIs), speech BCI has the potential to directly reflect auditory perception and thoughts, offering a promising communication alternative for patients with aphasia. Chinese is one of the most widely spoken languages in the world, whereas there is very limited research on speech BCIs for Chinese language. This paper reports a text-magnetoencephalography (MEG) dataset for non-invasive Chinese speech BCIs. It also proposes a multi-modality assisted speech decoding (MASD) algorithm to capture both text and acoustic information embedded in brain signals during speech activities. Experiment results demonstrated the effectiveness of both our text-MEG dataset and our proposed MASD algorithm. To our knowledge, this is the first study on modality-assisted decoding for non-invasive speech BCIs.
Numerous tools for neural network inference are currently available, yet many do not meet the requirements of real-time audio applications. In response, we introduce anira, an efficient cross-platform library. To ensure compatibility with a broad range of neural network architectures and frameworks, anira supports ONNX Runtime, LibTorch, and TensorFlow Lite as backends. Each inference engine exhibits real-time violations, which anira mitigates by decoupling the inference from the audio callback to a static thread pool. The library incorporates built-in latency management and extensive benchmarking capabilities, both crucial to ensure a continuous signal flow. Three different neural network architectures for audio effect emulation are then subjected to benchmarking across various configurations. Statistical modeling is employed to identify the influence of various factors on performance. The findings indicate that for stateless models, ONNX Runtime exhibits the lowest runtimes. For stateful models, LibTorch demonstrates the fastest performance. Our results also indicate that for certain model-engine combinations, the initial inferences take longer, particularly when these inferences exhibit a higher incidence of real-time violations.
Small-Footprint Keyword Spotting (SF-KWS) has gained popularity in today's landscape of smart voice-activated devices, smartphones, and Internet of Things (IoT) applications. This surge is attributed to the advancements in Deep Learning, enabling the identification of predefined words or keywords from a continuous stream of words. To implement the SF-KWS model on edge devices with low power and limited memory in real-world scenarios, a efficient Tiny Machine Learning (TinyML) framework is essential. In this study, we explore seven distinct categories of techniques namely, Model Architecture, Learning Techniques, Model Compression, Attention Awareness Architecture, Feature Optimization, Neural Network Search, and Hybrid Approaches, which are suitable for developing an SF-KWS system. This comprehensive overview will serve as a valuable resource for those looking to understand, utilize, or contribute to the field of SF-KWS. The analysis conducted in this work enables the identification of numerous potential research directions, encompassing insights from automatic speech recognition research and those specifically pertinent to the realm of spoken SF-KWS.
This paper investigates multi-agent reinforcement learning (MARL) in a partially observable, cooperative-competitive combat environment known as LAG. We describe the environment's setup, including agent actions, hierarchical controls, and reward design across different combat modes such as No Weapon and ShootMissile. Two representative algorithms are evaluated: HAPPO, an on-policy hierarchical variant of PPO, and HASAC, an off-policy method based on soft actor-critic. We analyze their training stability, reward progression, and inter-agent coordination capabilities. Experimental results show that HASAC performs well in simpler coordination tasks without weapons, while HAPPO demonstrates stronger adaptability in more dynamic and expressive scenarios involving missile combat. These findings provide insights into the trade-offs between on-policy and off-policy methods in multi-agent settings.
Motion Object Segmentation (MOS) is crucial for autonomous driving, as it enhances localization, path planning, map construction, scene flow estimation, and future state prediction. While existing methods achieve strong performance, balancing accuracy and real-time inference remains a challenge. To address this, we propose a logits-based knowledge distillation framework for MOS, aiming to improve accuracy while maintaining real-time efficiency. Specifically, we adopt a Bird's Eye View (BEV) projection-based model as the student and a non-projection model as the teacher. To handle the severe imbalance between moving and non-moving classes, we decouple them and apply tailored distillation strategies, allowing the teacher model to better learn key motion-related features. This approach significantly reduces false positives and false negatives. Additionally, we introduce dynamic upsampling, optimize the network architecture, and achieve a 7.69% reduction in parameter count, mitigating overfitting. Our method achieves a notable IoU of 78.8% on the hidden test set of the SemanticKITTI-MOS dataset and delivers competitive results on the Apollo dataset. The KDMOS implementation is available at https://github.com/SCNU-RISLAB/KDMOS.
We propose and study Hierarchical Ego Graph Neural Networks (HEGNNs), an expressive extension of graph neural networks (GNNs) with hierarchical node individualization, inspired by the Individualization-Refinement paradigm for graph isomorphism testing. HEGNNs generalize subgraph-GNNs and form a hierarchy of increasingly expressive models that, in the limit, can distinguish graphs up to isomorphism. We provide a logical characterization of HEGNN node classifiers, with and without subgraph restrictions, using graded hybrid logic. This characterization enables us to relate the separating power of HEGNNs to that of higher-order GNNs, GNNs enriched with local homomorphism count features, and color refinement algorithms based on Individualization-Refinement. Our experimental results confirm the practical feasibility of HEGNNs and show benefits in comparison with traditional GNN architectures, both with and without local homomorphism count features.
This survey provides a comprehensive overview of the emerging field of world models grounded in the foundation of acoustic physical information. It examines the theoretical underpinnings, essential methodological frameworks, and recent technological advancements in leveraging acoustic signals for high-fidelity environmental perception, causal physical reasoning, and predictive simulation of dynamic events. The survey explains how acoustic signals, as direct carriers of mechanical wave energy from physical events, encode rich, latent information about material properties, internal geometric structures, and complex interaction dynamics. Specifically, this survey establishes the theoretical foundation by explaining how fundamental physical laws govern the encoding of physical information within acoustic signals. It then reviews the core methodological pillars, including Physics-Informed Neural Networks (PINNs), generative models, and self-supervised multimodal learning frameworks. Furthermore, the survey details the significant applications of acoustic world models in robotics, autonomous driving, healthcare, and finance. Finally, it systematically outlines the important technical and ethical challenges while proposing a concrete roadmap for future research directions toward robust, causal, uncertainty-aware, and responsible acoustic intelligence. These elements collectively point to a research pathway towards embodied active acoustic intelligence, empowering AI systems to construct an internal "intuitive physics" engine through sound.
Many engineering and scientific fields have recently become interested in modeling terms in partial differential equations (PDEs) with neural networks. The resulting neural-network PDE model, being a function of the neural network parameters, can be calibrated to available data by optimizing over the PDE using gradient descent, where the gradient is evaluated in a computationally efficient manner by solving an adjoint PDE. These neural-network PDE models have emerged as an important research area in scientific machine learning. In this paper, we study the convergence of the adjoint gradient descent optimization method for training neural-network PDE models in the limit where both the number of hidden units and the training time tend to infinity. Specifically, for a general class of nonlinear parabolic PDEs with a neural network embedded in the source term, we prove convergence of the trained neural-network PDE solution to the target data (i.e., a global minimizer). The global convergence proof poses a unique mathematical challenge that is not encountered in finite-dimensional neural network convergence analyses due to (1) the neural network training dynamics involving a non-local neural network kernel operator in the infinite-width hidden layer limit where the kernel lacks a spectral gap for its eigenvalues and (2) the nonlinearity of the limit PDE system, which leads to a non-convex optimization problem, even in the infinite-width hidden layer limit (unlike in typical neual network training cases where the optimization problem becomes convex in the large neuron limit). The theoretical results are illustrated and empirically validated by numerical studies.
Accurate channel estimation is critical for high-performance Orthogonal Frequency-Division Multiplexing systems such as 5G New Radio, particularly under low signal-to-noise ratio and stringent latency constraints. This letter presents HELENA, a compact deep learning model that combines a lightweight convolutional backbone with two efficient attention mechanisms: patch-wise multi-head self-attention for capturing global dependencies and a squeeze-and-excitation block for local feature refinement. Compared to CEViT, a state-of-the-art vision transformer-based estimator, HELENA reduces inference time by 45.0\% (0.175\,ms vs.\ 0.318\,ms), achieves comparable accuracy ($-16.78$\,dB vs.\ $-17.30$\,dB), and requires $8\times$ fewer parameters (0.11M vs.\ 0.88M), demonstrating its suitability for low-latency, real-time deployment.
Existing work in automatic music generation has primarily focused on end-to-end systems that produce complete compositions or continuations. However, because musical composition is typically an iterative process, such systems make it difficult to engage in the back-and-forth between human and machine that is essential to computer-assisted creativity. In this study, we address the task of personalizable, multi-track, long-context, and controllable symbolic music infilling to enhance the process of computer-assisted composition. We present MIDI-RWKV, a novel model based on the RWKV-7 linear architecture, to enable efficient and coherent musical cocreation on edge devices. We also demonstrate that MIDI-RWKV admits an effective method of finetuning its initial state for personalization in the very-low-sample regime. We evaluate MIDI-RWKV and its state tuning on several quantitative and qualitative metrics, and release model weights and code at https://github.com/christianazinn/MIDI-RWKV.
While large language models have shown reasoning capabilities, their application to the audio modality, particularly in large audio-language models (ALMs), remains significantly underdeveloped. Addressing this gap requires a systematic approach, involving a capable base model, high-quality reasoning-oriented audio data, and effective training algorithms. In this study, we present a comprehensive solution: we introduce the Audio Logical Reasoning (ALR) dataset, consisting of 6,446 text-audio annotated samples specifically designed for complex reasoning tasks. Building on this resource, we propose SoundMind, a rule-based reinforcement learning (RL) algorithm tailored to endow ALMs with deep bimodal reasoning abilities. By training Qwen2.5-Omni-7B on the ALR dataset using SoundMind, our approach achieves state-of-the-art performance in audio logical reasoning. This work highlights the impact of combining high-quality, reasoning-focused datasets with specialized RL techniques, advancing the frontier of auditory intelligence in language models. Our code and the proposed dataset are available at https://github.com/xid32/SoundMind.
Designing effective reward functions remains a fundamental challenge in reinforcement learning (RL), as it often requires extensive human effort and domain expertise. While RL from human feedback has been successful in aligning agents with human intent, acquiring high-quality feedback is costly and labor-intensive, limiting its scalability. Recent advancements in foundation models present a promising alternative--leveraging AI-generated feedback to reduce reliance on human supervision in reward learning. Building on this paradigm, we introduce ERL-VLM, an enhanced rating-based RL method that effectively learns reward functions from AI feedback. Unlike prior methods that rely on pairwise comparisons, ERL-VLM queries large vision-language models (VLMs) for absolute ratings of individual trajectories, enabling more expressive feedback and improved sample efficiency. Additionally, we propose key enhancements to rating-based RL, addressing instability issues caused by data imbalance and noisy labels. Through extensive experiments across both low-level and high-level control tasks, we demonstrate that ERL-VLM significantly outperforms existing VLM-based reward generation methods. Our results demonstrate the potential of AI feedback for scaling RL with minimal human intervention, paving the way for more autonomous and efficient reward learning.
Zero-shot denoising aims to denoise observations without access to training samples or clean reference images. This setting is particularly relevant in practical imaging scenarios involving specialized domains such as medical imaging or biology. In this work, we propose the Zero-Shot Neural Compression Denoiser (ZS-NCD), a novel denoising framework based on neural compression. ZS-NCD treats a neural compression network as an untrained model, optimized directly on patches extracted from a single noisy image. The final reconstruction is then obtained by aggregating the outputs of the trained model over overlapping patches. Thanks to the built-in entropy constraints of compression architectures, our method naturally avoids overfitting and does not require manual regularization or early stopping. Through extensive experiments, we show that ZS-NCD achieves state-of-the-art performance among zero-shot denoisers for both Gaussian and Poisson noise, and generalizes well to both natural and non-natural images. Additionally, we provide new finite-sample theoretical results that characterize upper bounds on the achievable reconstruction error of general maximum-likelihood compression-based denoisers. These results further establish the theoretical foundations of compression-based denoising. Our code is available at: github.com/Computational-Imaging-RU/ZS-NCDenoiser.
This survey examines the rapidly evolving field of Deep Research systems -- AI-powered applications that automate complex research workflows through the integration of large language models, advanced information retrieval, and autonomous reasoning capabilities. We analyze more than 80 commercial and non-commercial implementations that have emerged since 2023, including OpenAI/Deep Research, Gemini/Deep Research, Perplexity/Deep Research, and numerous open-source alternatives. Through comprehensive examination, we propose a novel hierarchical taxonomy that categorizes systems according to four fundamental technical dimensions: foundation models and reasoning engines, tool utilization and environmental interaction, task planning and execution control, and knowledge synthesis and output generation. We explore the architectural patterns, implementation approaches, and domain-specific adaptations that characterize these systems across academic, scientific, business, and educational applications. Our analysis reveals both the significant capabilities of current implementations and the technical and ethical challenges they present regarding information accuracy, privacy, intellectual property, and accessibility. The survey concludes by identifying promising research directions in advanced reasoning architectures, multimodal integration, domain specialization, human-AI collaboration, and ecosystem standardization that will likely shape the future evolution of this transformative technology. By providing a comprehensive framework for understanding Deep Research systems, this survey contributes to both the theoretical understanding of AI-augmented knowledge work and the practical development of more capable, responsible, and accessible research technologies. The paper resources can be viewed at https://github.com/scienceaix/deepresearch.
Super-Resolution (SR) is a critical task in computer vision, focusing on reconstructing high-resolution (HR) images from low-resolution (LR) inputs. The field has seen significant progress through various challenges, particularly in single-image SR. Video Super-Resolution (VSR) extends this to the temporal domain, aiming to enhance video quality using methods like local, uni-, bi-directional propagation, or traditional upscaling followed by restoration. This challenge addresses VSR for conferencing, where LR videos are encoded with H.265 at fixed QPs. The goal is to upscale videos by a specific factor, providing HR outputs with enhanced perceptual quality under a low-delay scenario using causal models. The challenge included three tracks: general-purpose videos, talking head videos, and screen content videos, with separate datasets provided by the organizers for training, validation, and testing. We open-sourced a new screen content dataset for the SR task in this challenge. Submissions were evaluated through subjective tests using a crowdsourced implementation of the ITU-T Rec P.910.