Traditional approaches -- such as Win Probability Added (WPA)-based ranking or computer vision-driven event detection -- can identify scoring plays but often miss strategic depth, momentum shifts, and storyline progression. Manual curation remains the gold standard but is resource-intensive and not scalable. We introduce DIAMOND, an LLM-driven agent for context-aware baseball highlight summarization that integrates structured sports analytics with natural language reasoning. DIAMOND leverages sabermetric features -- Win Expectancy, WPA, and Leverage Index -- to quantify play importance, while an LLM module enhances selection based on contextual narrative value. This hybrid approach ensures both quantitative rigor and qualitative richness, surpassing the limitations of purely statistical or vision-based systems. Evaluated on five diverse Korean Baseball Organization League games, DIAMOND improves F1-score from 42.9% (WPA-only) to 84.8%, outperforming both commercial and statistical baselines. Though limited in scale, our results highlight the potential of modular, interpretable agent-based frameworks for event-level summarization in sports and beyond.
Recent research in deep learning-based Sound Event Detection (SED) has primarily focused on Convolutional Recurrent Neural Networks (CRNNs) and Transformer models. However, conventional 2D convolution-based models assume shift invariance along both the temporal and frequency axes, leadin to inconsistencies when dealing with frequency-dependent characteristics of acoustic signals. To address this issue, this study proposes Frequency Dynamic Convolution (FDY conv), which dynamically adjusts convolutional kernels based on the frequency composition of the input signal to enhance SED performance. FDY conv constructs an optimal frequency response by adaptively weighting multiple basis kernels based on frequency-specific attention weights. Experimental results show that applying FDY conv to CRNNs improves performance on the DESED dataset by 7.56% compared to the baseline CRNN. However, FDY conv has limitations in that it combines basis kernels of the same shape across all frequencies, restricting its ability to capture diverse frequency-specific characteristics. Additionally, the $3\times3$ basis kernel size is insufficient to capture a broader frequency range. To overcome these limitations, this study introduces an extended family of FDY conv models. Dilated FDY conv (DFD conv) applies convolutional kernels with various dilation rates to expand the receptive field along the frequency axis and enhance frequency-specific feature representation. Experimental results show that DFD conv improves performance by 9.27% over the baseline. Partial FDY conv (PFD conv) addresses the high computational cost of FDY conv, which results from performing all convolution operations with dynamic kernels. Since FDY conv may introduce unnecessary adaptivity for quasi-stationary sound events, PFD conv integrates standard 2D convolutions with frequency-adaptive kernels to reduce computational complexity while maintaining performance. Experimental results demonstrate that PFD conv improves performance by 7.80% over the baseline while reducing the number of parameters by 54.4% compared to FDY conv. Multi-Dilated FDY conv (MDFD conv) extends DFD conv by addressing its structural limitation of applying the same dilation across all frequencies. By utilizing multiple convolutional kernels with different dilation rates, MDFD conv effectively captures diverse frequency-dependent patterns. Experimental results indicate that MDFD conv achieves the highest performance, improving the baseline CRNN performance by 10.98%. Furthermore, standard FDY conv employs Temporal Average Pooling, which assigns equal weight to all frames along the time axis, limiting its ability to effectively capture transient events. To overcome this, this study proposes TAP-FDY conv (TFD conv), which integrates Temporal Attention Pooling (TA) that focuses on salient features, Velocity Attention Pooling (VA) that emphasizes transient characteristics, and Average Pooling (AP) that captures stationary properties. TAP-FDY conv achieves the same performance as MDFD conv but reduces the number of parameters by approximately 30.01% (12.703M vs. 18.157M), achieving equivalent accuracy with lower computational complexity. Class-wise performance analysis reveals that FDY conv improves detection of non-stationary events, DFD conv is particularly effective for events with broad spectral features, and PFD conv enhances the detection of quasi-stationary events. Additionally, TFD conv (TFD-CRNN) demonstrates strong performance in detecting transient events. In the case studies, PFD conv effectively captures stable signal patterns in tank powertrain fault recognition, DFD conv recognizes wide harmonic spectral patterns on speed-varying motor fault recognition, while TFD conv outperforms other models in detecting transient signals in offshore arc detection. These results suggest that frequency-adaptive convolutions and their extended variants provide a robust alternative to conventional 2D convolutions in deep learning-based audio processing.
Accurate modeling of infrasound transmission loss is essential for evaluating the performance of the International Monitoring System, enabling the effective design and maintenance of infrasound stations to support compliance of the Comprehensive Nuclear-Test-Ban Treaty. State-of-the-art propagation modeling tools enable transmission loss to be finely simulated using atmospheric models. However, the computational cost prohibits the exploration of a large parameter space in operational monitoring applications. To address this, recent studies made use of a deep learning algorithm capable of making transmission loss predictions almost instantaneously. However, the use of nudged atmospheric models leads to an incomplete representation of the medium, and the absence of temperature as an input makes the algorithm incompatible with long range propagation. In this study, we address these limitations by using both wind and temperature fields as inputs to a neural network, simulated up to 130 km altitude and 4,000 km distance. We also optimize several aspects of the neural network architecture. We exploit convolutional and recurrent layers to capture spatially and range-dependent features embedded in realistic atmospheric models, improving the overall performance. The neural network reaches an average error of 4 dB compared to full parabolic equation simulations and provides epistemic and data-related uncertainty estimates. Its evaluation on the 2022 Hunga Tonga-Hunga Ha'apai volcanic eruption demonstrates its prediction capability using atmospheric conditions and frequencies not included in the training. This represents a significant step towards near real-time assessment of International Monitoring System detection thresholds of explosive sources.
Zero-shot Event Detection (ED), the task of identifying event mentions in natural language text without any training data, is critical for document understanding in specialized domains. Understanding the complex event ontology, extracting domain-specific triggers from the passage, and structuring them appropriately overloads and limits the utility of Large Language Models (LLMs) for zero-shot ED. To this end, we propose DiCoRe, a divergent-convergent reasoning framework that decouples the task of ED using Dreamer and Grounder. Dreamer encourages divergent reasoning through open-ended event discovery, which helps to boost event coverage. Conversely, Grounder introduces convergent reasoning to align the free-form predictions with the task-specific instructions using finite-state machine guided constrained decoding. Additionally, an LLM-Judge verifies the final outputs to ensure high precision. Through extensive experiments on six datasets across five domains and nine LLMs, we demonstrate how DiCoRe consistently outperforms prior zero-shot, transfer-learning, and reasoning baselines, achieving 4-7% average F1 gains over the best baseline -- establishing DiCoRe as a strong zero-shot ED framework.
Bioacoustic sound event detection (BioSED) is crucial for biodiversity conservation but faces practical challenges during model development and training: limited amounts of annotated data, sparse events, species diversity, and class imbalance. To address these challenges efficiently with a limited labeling budget, we apply the mismatch-first farthest-traversal (MFFT), an active learning method integrating committee voting disagreement and diversity analysis. We also refine an existing BioSED dataset specifically for evaluating active learning algorithms. Experimental results demonstrate that MFFT achieves a mAP of 68% when cold-starting and 71% when warm-starting (which is close to the fully-supervised mAP of 75%) while using only 2.3% of the annotations. Notably, MFFT excels in cold-start scenarios and with rare species, which are critical for monitoring endangered species, demonstrating its practical value.
In our prior works, we introduced a scalable streaming speech synthesis model, CosyVoice 2, which integrates a large language model (LLM) and a chunk-aware flow matching (FM) model, and achieves low-latency bi-streaming speech synthesis and human-parity quality. Despite these advancements, CosyVoice 2 exhibits limitations in language coverage, domain diversity, data volume, text formats, and post-training techniques. In this paper, we present CosyVoice 3, an improved model designed for zero-shot multilingual speech synthesis in the wild, surpassing its predecessor in content consistency, speaker similarity, and prosody naturalness. Key features of CosyVoice 3 include: 1) A novel speech tokenizer to improve prosody naturalness, developed via supervised multi-task training, including automatic speech recognition, speech emotion recognition, language identification, audio event detection, and speaker analysis. 2) A new differentiable reward model for post-training applicable not only to CosyVoice 3 but also to other LLM-based speech synthesis models. 3) Dataset Size Scaling: Training data is expanded from ten thousand hours to one million hours, encompassing 9 languages and 18 Chinese dialects across various domains and text formats. 4) Model Size Scaling: Model parameters are increased from 0.5 billion to 1.5 billion, resulting in enhanced performance on our multilingual benchmark due to the larger model capacity. These advancements contribute significantly to the progress of speech synthesis in the wild. We encourage readers to listen to the demo at https://funaudiollm.github.io/cosyvoice3.
Self-supervised learning (SSL) models offer powerful representations for sound event detection (SED), yet their synergistic potential remains underexplored. This study systematically evaluates state-of-the-art SSL models to guide optimal model selection and integration for SED. We propose a framework that combines heterogeneous SSL representations (e.g., BEATs, HuBERT, WavLM) through three fusion strategies: individual SSL embedding integration, dual-modal fusion, and full aggregation. Experiments on the DCASE 2023 Task 4 Challenge reveal that dual-modal fusion (e.g., CRNN+BEATs+WavLM) achieves complementary performance gains, while CRNN+BEATs alone delivers the best results among individual SSL models. We further introduce normalized sound event bounding boxes (nSEBBs), an adaptive post-processing method that dynamically adjusts event boundary predictions, improving PSDS1 by up to 4% for standalone SSL models. These findings highlight the compatibility and complementarity of SSL architectures, providing guidance for task-specific fusion and robust SED system design.
In this paper, we study the challenges of detecting events on social media, where traditional unimodal systems struggle due to the rapid and multimodal nature of data dissemination. We employ a range of models, including unimodal ModernBERT and ConvNeXt-V2, multimodal fusion techniques, and advanced generative models like GPT-4o, and LLaVA. Additionally, we also study the effect of providing multimodal generative models (such as GPT-4o) with a single modality to assess their efficacy. Our results indicate that while multimodal approaches notably outperform unimodal counterparts, generative approaches despite having a large number of parameters, lag behind supervised methods in precision. Furthermore, we also found that they lag behind instruction-tuned models because of their inability to generate event classes correctly. During our error analysis, we discovered that common social media issues such as leet speak, text elongation, etc. are effectively handled by generative approaches but are hard to tackle using supervised approaches.
In this work, we enable gamers to share their gaming experience on social media by automatically generating eye-catching highlight reels from their gameplay session Our automation will save time for gamers while increasing audience engagement. We approach the highlight generation problem by first identifying intervals in the video where interesting events occur and then concatenate them. We developed an in-house gameplay event detection dataset containing interesting events annotated by humans using VIA video annotator. Traditional techniques for highlight detection such as game engine integration requires expensive collaboration with game developers. OCR techniques which detect patches of specific images or texts require expensive per game engineering and may not generalize across game UI and different language. We finetuned a multimodal general purpose video understanding model such as X-CLIP using our dataset which generalizes across multiple games in a genre without per game engineering. Prompt engineering was performed to improve the classification performance of this multimodal model. Our evaluation showed that such a finetuned model can detect interesting events in first person shooting games from unseen gameplay footage with more than 90% accuracy. Moreover, our model performed significantly better on low resource games (small dataset) when trained along with high resource games, showing signs of transfer learning. To make the model production ready, we used ONNX libraries to enable cross platform inference. These libraries also provide post training quantization tools to reduce model size and inference time for deployment. ONNX runtime libraries with DirectML backend were used to perform efficient inference on Windows OS. We show that natural language supervision in the X-CLIP model leads to data efficient and highly performant video recognition models.
Event-based object detection has gained increasing attention due to its advantages such as high temporal resolution, wide dynamic range, and asynchronous address-event representation. Leveraging these advantages, Spiking Neural Networks (SNNs) have emerged as a promising approach, offering low energy consumption and rich spatiotemporal dynamics. To further enhance the performance of event-based object detection, this study proposes a novel hybrid spike vision Transformer (HsVT) model. The HsVT model integrates a spatial feature extraction module to capture local and global features, and a temporal feature extraction module to model time dependencies and long-term patterns in event sequences. This combination enables HsVT to capture spatiotemporal features, improving its capability to handle complex event-based object detection tasks. To support research in this area, we developed and publicly released The Fall Detection Dataset as a benchmark for event-based object detection tasks. This dataset, captured using an event-based camera, ensures facial privacy protection and reduces memory usage due to the event representation format. We evaluated the HsVT model on GEN1 and Fall Detection datasets across various model sizes. Experimental results demonstrate that HsVT achieves significant performance improvements in event detection with fewer parameters.
Video event detection has become an essential component of sports analytics, enabling automated identification of key moments and enhancing performance analysis, viewer engagement, and broadcast efficiency. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs) and Transformers, have significantly improved accuracy and efficiency in Temporal Action Localization (TAL), Action Spotting (AS), and Precise Event Spotting (PES). This survey provides a comprehensive overview of these three key tasks, emphasizing their differences, applications, and the evolution of methodological approaches. We thoroughly review and categorize existing datasets and evaluation metrics specifically tailored for sports contexts, highlighting the strengths and limitations of each. Furthermore, we analyze state-of-the-art techniques, including multi-modal approaches that integrate audio and visual information, methods utilizing self-supervised learning and knowledge distillation, and approaches aimed at generalizing across multiple sports. Finally, we discuss critical open challenges and outline promising research directions toward developing more generalized, efficient, and robust event detection frameworks applicable to diverse sports. This survey serves as a foundation for future research on efficient, generalizable, and multi-modal sports event detection.
Analyzing multi-featured time series data is critical for space missions making efficient event detection, potentially onboard, essential for automatic analysis. However, limited onboard computational resources and data downlink constraints necessitate robust methods for identifying regions of interest in real time. This work presents an adaptive outlier detection algorithm based on the reconstruction error of Principal Component Analysis (PCA) for feature reduction, designed explicitly for space mission applications. The algorithm adapts dynamically to evolving data distributions by using Incremental PCA, enabling deployment without a predefined model for all possible conditions. A pre-scaling process normalizes each feature's magnitude while preserving relative variance within feature types. We demonstrate the algorithm's effectiveness in detecting space plasma events, such as distinct space environments, dayside and nightside transients phenomena, and transition layers through NASA's MMS mission observations. Additionally, we apply the method to NASA's THEMIS data, successfully identifying a dayside transient using onboard-available measurements.
The World Health Organization suggests that road traffic crashes cost approximately 518 billion dollars globally each year, which accounts for 3% of the gross domestic product for most countries. Most fatal road accidents in urban areas involve Vulnerable Road Users (VRUs). Smart cities environments present innovative approaches to combat accidents involving cutting-edge technologies, that include advanced sensors, extensive datasets, Machine Learning (ML) models, communication systems, and edge computing. This paper proposes a strategy and an implementation of a system for road monitoring and safety for smart cities, based on Computer Vision (CV) and edge computing. Promising results were obtained by implementing vision algorithms and tracking using surveillance cameras, that are part of a Smart City testbed, the Aveiro Tech City Living Lab (ATCLL). The algorithm accurately detects and tracks cars, pedestrians, and bicycles, while predicting the road state, the distance between moving objects, and inferring on collision events to prevent collisions, in near real-time.
Purpose: The operating room (OR) is a complex environment where optimizing workflows is critical to reduce costs and improve patient outcomes. The use of computer vision approaches for the automatic recognition of perioperative events enables identification of bottlenecks for OR optimization. However, privacy concerns limit the use of computer vision for automated event detection from OR videos, which makes privacy-preserving approaches needed for OR workflow analysis. Methods: We propose a two-stage pipeline for privacy-preserving OR video analysis and event detection. In the first stage, we leverage vision foundation models for depth estimation and semantic segmentation to generate de-identified Digital Twins (DT) of the OR from conventional RGB videos. In the second stage, we employ the SafeOR model, a fused two-stream approach that processes segmentation masks and depth maps for OR event detection. We evaluate this method on an internal dataset of 38 simulated surgical trials with five event classes. Results: Our results indicate that this DT-based approach to the OR event detection model achieves performance on par and sometimes even better than raw RGB video-based models on detecting OR events. Conclusion: DTs enable privacy-preserving OR workflow analysis, facilitating the sharing of de-identified data across institutions and they can potentially enhance model generalizability by mitigating domain-specific appearance differences.
Sound event localization and detection (SELD) is a task for the classification of sound events and the identification of direction of arrival (DoA) utilizing multichannel acoustic signals. For effective classification and localization, a channel-spectro-temporal transformer (CST-former) was suggested. CST-former employs multidimensional attention mechanisms across the spatial, spectral, and temporal domains to enlarge the model's capacity to learn the domain information essential for event detection and DoA estimation over time. In this work, we present an enhanced version of CST-former with multiscale unfolded local embedding (MSULE) developed to capture and aggregate domain information over multiple time-frequency scales. Also, we propose finetuning and post-processing techniques beneficial for conducting the SELD task over limited training datasets. In-depth ablation studies of the proposed architecture and detailed analysis on the proposed modules are carried out to validate the efficacy of multidimensional attentions on the SELD task. Empirical validation through experimentation on STARSS22 and STARSS23 datasets demonstrates the remarkable performance of CST-former and post-processing techniques without using external data.
Recent advances in deep learning, particularly frequency dynamic convolution (FDY conv), have significantly improved sound event detection (SED) by enabling frequency-adaptive feature extraction. However, FDY conv relies on temporal average pooling, which treats all temporal frames equally, limiting its ability to capture transient sound events such as alarm bells, door knocks, and speech plosives. To address this limitation, we propose temporal attention pooling frequency dynamic convolution (TFD conv) to replace temporal average pooling with temporal attention pooling (TAP). TAP adaptively weights temporal features through three complementary mechanisms: time attention pooling (TA) for emphasizing salient features, velocity attention pooling (VA) for capturing transient changes, and conventional average pooling for robustness to stationary signals. Ablation studies show that TFD conv improves average PSDS1 by 3.02% over FDY conv with only a 14.8% increase in parameter count. Classwise ANOVA and Tukey HSD analysis further demonstrate that TFD conv significantly enhances detection performance for transient-heavy events, outperforming existing FDY conv models. Notably, TFD conv achieves a maximum PSDS1 score of 0.456, surpassing previous state-of-the-art SED systems. We also explore the compatibility of TAP with other FDY conv variants, including dilated FDY conv (DFD conv), partial FDY conv (PFD conv), and multi-dilated FDY conv (MDFD conv). Among these, the integration of TAP with MDFD conv achieves the best result with a PSDS1 score of 0.459, validating the complementary strengths of temporal attention and multi-scale frequency adaptation. These findings establish TFD conv as a powerful and generalizable framework for enhancing both transient sensitivity and overall feature robustness in SED.
Analyzing Fast, Frequent, and Fine-grained (F$^3$) events presents a significant challenge in video analytics and multi-modal LLMs. Current methods struggle to identify events that satisfy all the F$^3$ criteria with high accuracy due to challenges such as motion blur and subtle visual discrepancies. To advance research in video understanding, we introduce F$^3$Set, a benchmark that consists of video datasets for precise F$^3$ event detection. Datasets in F$^3$Set are characterized by their extensive scale and comprehensive detail, usually encompassing over 1,000 event types with precise timestamps and supporting multi-level granularity. Currently, F$^3$Set contains several sports datasets, and this framework may be extended to other applications as well. We evaluated popular temporal action understanding methods on F$^3$Set, revealing substantial challenges for existing techniques. Additionally, we propose a new method, F$^3$ED, for F$^3$ event detections, achieving superior performance. The dataset, model, and benchmark code are available at https://github.com/F3Set/F3Set.
Spiking neural networks (SNN) hold the promise of being a more biologically plausible, low-energy alternative to conventional artificial neural networks. Their time-variant nature makes them particularly suitable for processing time-resolved, sparse binary data. In this paper, we investigate the potential of leveraging SNNs for the detection of photon coincidences in positron emission tomography (PET) data. PET is a medical imaging technique based on injecting a patient with a radioactive tracer and detecting the emitted photons. One central post-processing task for inferring an image of the tracer distribution is the filtering of invalid hits occurring due to e.g. absorption or scattering processes. Our approach, coined PETNet, interprets the detector hits as a binary-valued spike train and learns to identify photon coincidence pairs in a supervised manner. We introduce a dedicated multi-objective loss function and demonstrate the effects of explicitly modeling the detector geometry on simulation data for two use-cases. Our results show that PETNet can outperform the state-of-the-art classical algorithm with a maximal coincidence detection $F_1$ of 95.2%. At the same time, PETNet is able to predict photon coincidences up to 36 times faster than the classical approach, highlighting the great potential of SNNs in particle physics applications.
Translocation event detection from raw nanopore current signals is a fundamental step in nanopore signal analysis. Traditional data analysis methods rely on user-defined parameters to extract event information, making the interpretation of experimental results sensitive to parameter choice. While Machine Learning (ML) has seen widespread adoption across various scientific fields, its potential remains underexplored in solid-state nanopore research. In this work, we introduce a nanopore signal generator capable of producing extensive synthetic datasets for machine learning applications and benchmarking nanopore signal analysis platforms. Using this generator, we train deep learning models to detect translocation events directly from raw signals, achieving over 99% true event detection with minimal false positives.
In this paper, we propose a novel formula-driven supervised learning (FDSL) framework for pre-training an environmental sound analysis model by leveraging acoustic signals parametrically synthesized through formula-driven methods. Specifically, we outline detailed procedures and evaluate their effectiveness for sound event detection (SED). The SED task, which involves estimating the types and timings of sound events, is particularly challenged by the difficulty of acquiring a sufficient quantity of accurately labeled training data. Moreover, it is well known that manually annotated labels often contain noises and are significantly influenced by the subjective judgment of annotators. To address these challenges, we propose a novel pre-training method that utilizes a synthetic dataset, Formula-SED, where acoustic data are generated solely based on mathematical formulas. The proposed method enables large-scale pre-training by using the synthesis parameters applied at each time step as ground truth labels, thereby eliminating label noise and bias. We demonstrate that large-scale pre-training with Formula-SED significantly enhances model accuracy and accelerates training, as evidenced by our results in the DESED dataset used for DCASE2023 Challenge Task 4. The project page is at https://yutoshibata07.github.io/Formula-SED/