Xin Gu

Yaojie Shen, Xin Gu, Kai Xu et al.

Existing video captioning approaches typically require to first sample video frames from a decoded video and then conduct a subsequent process (e.g., feature extraction and/or captioning model learning). In this pipeline, manual frame sampling may ignore key information in videos and thus degrade performance. Additionally, redundant information in the sampled frames may result in low efficiency in the inference of video captioning. Addressing this, we study video captioning from a different perspective in compressed domain, which brings multi-fold advantages over the existing pipeline: 1) Compared to raw images from the decoded video, the compressed video, consisting of I-frames, motion vectors and residuals, is highly distinguishable, which allows us to leverage the entire video for learning without manual sampling through a specialized model design; 2) The captioning model is more efficient in inference as smaller and less redundant information is processed. We propose a simple yet effective end-to-end transformer in the compressed domain for video captioning that enables learning from the compressed video for captioning. We show that even with a simple design, our method can achieve state-of-the-art performance on different benchmarks while running almost 2x faster than existing approaches. Code is available at https://github.com/acherstyx/CoCap.

Libo Zhang, Xin Gu, Congcong Li et al.

Generic event boundary detection aims to localize the generic, taxonomy-free event boundaries that segment videos into chunks. Existing methods typically require video frames to be decoded before feeding into the network, which contains significant spatio-temporal redundancy and demands considerable computational power and storage space. To remedy these issues, we propose a novel compressed video representation learning method for event boundary detection that is fully end-to-end leveraging rich information in the compressed domain, i.e., RGB, motion vectors, residuals, and the internal group of pictures (GOP) structure, without fully decoding the video. Specifically, we use lightweight ConvNets to extract features of the P-frames in the GOPs and spatial-channel attention module (SCAM) is designed to refine the feature representations of the P-frames based on the compressed information with bidirectional information flow. To learn a suitable representation for boundary detection, we construct the local frames bag for each candidate frame and use the long short-term memory (LSTM) module to capture temporal relationships. We then compute frame differences with group similarities in the temporal domain. This module is only applied within a local window, which is critical for event boundary detection. Finally a simple classifier is used to determine the event boundaries of video sequences based on the learned feature representation. To remedy the ambiguities of annotations and speed up the training process, we use the Gaussian kernel to preprocess the ground-truth event boundaries. Extensive experiments conducted on the Kinetics-GEBD and TAPOS datasets demonstrate that the proposed method achieves considerable improvements compared to previous end-to-end approach while running at the same speed. The code is available at https://github.com/GX77/LCVSL.

Xin Gu, Guang Chen, Yufei Wang et al.

Video captioning aims to describe the content of videos using natural language. Although significant progress has been made, there is still much room to improve the performance for real-world applications, mainly due to the long-tail words challenge. In this paper, we propose a text with knowledge graph augmented transformer (TextKG) for video captioning. Notably, TextKG is a two-stream transformer, formed by the external stream and internal stream. The external stream is designed to absorb additional knowledge, which models the interactions between the additional knowledge, e.g., pre-built knowledge graph, and the built-in information of videos, e.g., the salient object regions, speech transcripts, and video captions, to mitigate the long-tail words challenge. Meanwhile, the internal stream is designed to exploit the multi-modality information in videos (e.g., the appearance of video frames, speech transcripts, and video captions) to ensure the quality of caption results. In addition, the cross attention mechanism is also used in between the two streams for sharing information. In this way, the two streams can help each other for more accurate results. Extensive experiments conducted on four challenging video captioning datasets, i.e., YouCookII, ActivityNet Captions, MSRVTT, and MSVD, demonstrate that the proposed method performs favorably against the state-of-the-art methods. Specifically, the proposed TextKG method outperforms the best published results by improving 18.7% absolute CIDEr scores on the YouCookII dataset.

Xin Gu, Gautam Kamath, Zhiwei Steven Wu

Differentially private stochastic gradient descent privatizes model training by injecting noise into each iteration, where the noise magnitude increases with the number of model parameters. Recent works suggest that we can reduce the noise by leveraging public data for private machine learning, by projecting gradients onto a subspace prescribed by the public data. However, given a choice of public datasets, it is not a priori clear which one may be most appropriate for the private task. We give an algorithm for selecting a public dataset by measuring a low-dimensional subspace distance between gradients of the public and private examples. We provide theoretical analysis demonstrating that the excess risk scales with this subspace distance. This distance is easy to compute and robust to modifications in the setting. Empirical evaluation shows that trained model accuracy is monotone in this distance.

Renteng Yuan, Mohamed Abdel-Aty, Xin Gu et al.

Accurately detecting and predicting lane change (LC)processes of human-driven vehicles can help autonomous vehicles better understand their surrounding environment, recognize potential safety hazards, and improve traffic safety. This paper focuses on LC processes, first developing a temporal convolutional network with an attention mechanism (TCN-ATM) model to recognize LC intention. Considering the intrinsic relationship among output variables, the Multi-task Learning (MTL)framework is employed to simultaneously predict multiple LC vehicle status indicators. Furthermore, a unified modeling framework for LC intention recognition and driving status prediction (LC-IR-SP) is developed. The results indicate that the classification accuracy of LC intention was improved from 96.14% to 98.20% when incorporating the attention mechanism into the TCN model. For LC vehicle status prediction issues, three multi-tasking learning models are constructed based on MTL framework. The results indicate that the MTL-LSTM model outperforms the MTL-TCN and MTL-TCN-ATM models. Compared to the corresponding single-task model, the MTL-LSTM model demonstrates an average decrease of 26.04% in MAE and 25.19% in RMSE.

Xin Gu, Hanhua Ye, Guang Chen et al.

This paper describes our champion solution for the CVPR2022 Generic Event Boundary Captioning (GEBC) competition. GEBC requires the captioning model to have a comprehension of instantaneous status changes around the given video boundary, which makes it much more challenging than conventional video captioning task. In this paper, a Dual-Stream Transformer with improvements on both video content encoding and captions generation is proposed: (1) We utilize three pre-trained models to extract the video features from different granularities. Moreover, we exploit the types of boundary as hints to help the model generate captions. (2) We particularly design an model, termed as Dual-Stream Transformer, to learn discriminative representations for boundary captioning. (3) Towards generating content-relevant and human-like captions, we improve the description quality by designing a word-level ensemble strategy. The promising results on the GEBC test split demonstrate the efficacy of our proposed model.

Shaleeza Sohail, Zongwen Fan, Xin Gu et al.

In recent years cybersecurity has become a major concern in adaptation of smart applications. Specially, in smart homes where a large number of IoT devices are used having a secure and trusted mechanisms can provide peace of mind for users. Accurate detection of cyber attacks is crucial, however precise identification of the type of attacks plays a huge role if devising the countermeasure for protecting the system. Artificial Neural Networks (ANN) have provided promising results for detecting any security attacks for smart applications. However, due to complex nature of the model used for this technique it is not easy for normal users to trust ANN based security solutions. Also, selection of right hyperparameters for ANN architecture plays a crucial role in the accurate detection of security attacks, especially when it come to identifying the subcategories of attacks. In this paper, we propose a model that considers both the issues of explainability of ANN model and the hyperparameter selection for this approach to be easily trusted and adapted by users of smart home applications. Also, our approach considers a subset of the dataset for optimal selection of hyperparamters to reduce the overhead of the process of ANN architecture design. Distinctively this paper focuses on configuration, performance and evaluation of ANN architecture for identification of five categorical attacks and nine subcategorical attacks. Using a very recent IoT dataset our approach showed high performance for intrusion detection with 99.9%, 99.7%, and 97.7% accuracy for Binary, Category, and Subcategory level classification of attacks.

Su Yeon Han, Yubin Lee, Jooyoung Yoo et al.

The increasing frequency and severity of wildfire in California, exacerbated by prolonged drought and environmental changes, pose significant challenges to urban community resilience and equitable emergency response. The study investigates issues of accessibility to shelters during the Palisades and Eaton Fires which started in January 2025 in Southern California that led to over 180,000 displacements and the loss of 16,000 structures. Despite coordinated efforts of many organizations' emergency assistance, shelter shortages left many evacuees without safety or accessible refuge. This research aims to measure shelter accessibility during the fires' peak, evaluate whether existing shelter capacity met the demand, and identify spatial disparities in access. Findings reveal severe shelter shortages and pronounced inequities in access to shelters, particularly in geographically isolated regions and mountainous areas. To address these challenges, we implemented shelter placement strategies using both capacity-based and distance-based approaches, demonstrating potential improvements in accessibility and equity. The findings underscore the critical need for strategic shelter planning and infrastructure development to enhance disaster readiness and reduce vulnerability in regions that frequently experience wildfires.

Xin Gu, Heng Fan, Yan Huang et al.

Spatio-temporal video grounding (or STVG) task aims at locating a spatio-temporal tube for a specific instance given a text query. Despite advancements, current methods easily suffer the distractors or heavy object appearance variations in videos due to insufficient object information from the text, leading to degradation. Addressing this, we propose a novel framework, context-guided STVG (CG-STVG), which mines discriminative instance context for object in videos and applies it as a supplementary guidance for target localization. The key of CG-STVG lies in two specially designed modules, including instance context generation (ICG), which focuses on discovering visual context information (in both appearance and motion) of the instance, and instance context refinement (ICR), which aims to improve the instance context from ICG by eliminating irrelevant or even harmful information from the context. During grounding, ICG, together with ICR, are deployed at each decoding stage of a Transformer architecture for instance context learning. Particularly, instance context learned from one decoding stage is fed to the next stage, and leveraged as a guidance containing rich and discriminative object feature to enhance the target-awareness in decoding feature, which conversely benefits generating better new instance context for improving localization finally. Compared to existing methods, CG-STVG enjoys object information in text query and guidance from mined instance visual context for more accurate target localization. In our experiments on three benchmarks, including HCSTVG-v1/-v2 and VidSTG, CG-STVG sets new state-of-the-arts in m_tIoU and m_vIoU on all of them, showing its efficacy. The code will be released at https://github.com/HengLan/CGSTVG.

Xin Gu, Ming Li, Libo Zhang et al.

High-quality training triplets (instruction, original image, edited image) are essential for instruction-based image editing. Predominant training datasets (e.g., InsPix2Pix) are created using text-to-image generative models (e.g., Stable Diffusion, DALL-E) which are not trained for image editing. Accordingly, these datasets suffer from inaccurate instruction following, poor detail preserving, and generation artifacts. In this paper, we propose to address the training data quality issue with multi-perspective reward data instead of refining the ground-truth image quality. 1) we first design a quantitative metric system based on best-in-class LVLM (Large Vision Language Model), i.e., GPT-4o in our case, to evaluate the generation quality from 3 perspectives, namely, instruction following, detail preserving, and generation quality. For each perspective, we collected quantitative score in $0\sim 5$ and text descriptive feedback on the specific failure points in ground-truth edited images, resulting in a high-quality editing reward dataset, i.e., RewardEdit20K. 2) We further proposed a novel training framework to seamlessly integrate the metric output, regarded as multi-reward, into editing models to learn from the imperfect training triplets. During training, the reward scores and text descriptions are encoded as embeddings and fed into both the latent space and the U-Net of the editing models as auxiliary conditions. 3) We also build a challenging evaluation benchmark with real-world images/photos and diverse editing instructions, named Real-Edit. Experiments indicate that our multi-reward conditioned model outperforms its no-reward counterpart on two popular editing pipelines, i.e., InsPix2Pix and SmartEdit. Code is released at https://github.com/bytedance/Multi-Reward-Editing.

Haoji Zhang, Xin Gu, Jiawen Li et al.

The video reasoning ability of multimodal large language models (MLLMs) is crucial for downstream tasks like video question answering and temporal grounding. While recent approaches have explored text-based chain-of-thought (CoT) reasoning for MLLMs, these methods often suffer from limited cross-modal interaction and increased hallucination, especially with longer videos or reasoning chains. To address these challenges, we propose Video Intelligence via Tool-Augmented Learning (VITAL), a novel end-to-end agentic video reasoning framework. With a visual toolbox, the model can densely sample new video frames on demand and generate multimodal CoT for precise long video reasoning. We observe that temporal grounding and question answering are mutually beneficial for video understanding tasks. Therefore, we construct two high-quality multi-task video reasoning datasets MTVR-CoT-72k for supervised fine-tuning and MTVR-RL-110k for reinforcement learning. Moreover, we propose a Difficulty-aware Group Relative Policy Optimization algorithm (DGRPO) to mitigate difficulty imbalance in multi-task reinforcement learning. Extensive experiments on 11 challenging video understanding benchmarks demonstrate the advanced reasoning ability of VITAL, outperforming existing methods in video question answering and temporal grounding tasks, especially in long video scenarios. Code is available at https://zhang9302002.github.io/thinkingwithvideos-page/.

Xin Gu, Libo Zhang, Fan Chen et al.

This paper focuses on understanding the predominant video creation pipeline, i.e., compositional video editing with six main types of editing components, including video effects, animation, transition, filter, sticker, and text. In contrast to existing visual representation learning of visual materials (i.e., images/videos), we aim to learn visual representations of editing actions/components that are generally applied on raw materials. We start by proposing the first large-scale dataset for editing components of video creation, which covers about $3,094$ editing components with $618,800$ videos. Each video in our dataset is rendered by various image/video materials with a single editing component, which supports atomic visual understanding of different editing components. It can also benefit several downstream tasks, e.g., editing component recommendation, editing component recognition/retrieval, etc. Existing visual representation methods perform poorly because it is difficult to disentangle the visual appearance of editing components from raw materials. To that end, we benchmark popular alternative solutions and propose a novel method that learns to attend to the appearance of editing components regardless of raw materials. Our method achieves favorable results on editing component retrieval/recognition compared to the alternative solutions. A user study is also conducted to show that our representations cluster visually similar editing components better than other alternatives. Furthermore, our learned representations used to transition recommendation tasks achieve state-of-the-art results on the AutoTransition dataset. The code and dataset are available at https://github.com/GX77/Edit3K .

Xin Gu, Haoji Zhang, Qihang Fan et al.

Spatio-temporal video grounding (STVG) requires localizing a target object in untrimmed videos both temporally and spatially from natural language descriptions. Despite their strong language understanding, multimodal large language models (MLLMs) underperform on STVG due to misaligned training objectives and weak fine-grained region-word alignment in standard visual encoders. To address this, we propose STVG-o1, the first framework that enables off-the-shelf MLLMs to achieve state-of-the-art STVG performance without any architectural modifications. Our method introduces a bounding-box chain-of-thought mechanism that explicitly reasons about spatio-temporal locations in an intermediate step before producing the final prediction. We further design a multi-dimensional reinforcement reward function consisting of format, consistency, temporal, spatial, and think rewards, which provides geometry-aware supervision through reinforcement fine-tuning. Evaluated on HCSTVG-v1/v2 and VidSTG, STVG-o1 sets new state-of-the-art results on HCSTVG, outperforming the best task-specific method by 7.3\% m\_tIoU on HCSTVG-v1, matching specialized models on VidSTG, and surpassing all existing MLLM-based approaches by large margins. It also demonstrates strong open-vocabulary generalization across datasets, establishing MLLMs as viable and powerful backbones for precise spatio-temporal grounding. Our code and models will be released.

Xin Gu, Bing Fan, Jiali Yao et al.

In real scenarios, videos can span several minutes or even hours. However, existing research on spatio-temporal video grounding (STVG), given a textual query, mainly focuses on localizing targets in short videos of tens of seconds, typically less than one minute, which limits real-world applications. In this paper, we explore Long-Form STVG (LF-STVG), which aims to locate targets in long-term videos. Compared with short videos, long-term videos contain much longer temporal spans and more irrelevant information, making it difficult for existing STVG methods that process all frames at once. To address this challenge, we propose an AutoRegressive Transformer architecture for LF-STVG, termed ART-STVG. Unlike conventional STVG methods that require the entire video sequence to make predictions at once, ART-STVG treats the video as streaming input and processes frames sequentially, enabling efficient handling of long videos. To model spatio-temporal context, we design spatial and temporal memory banks and apply them to the decoders. Since memories from different moments are not always relevant to the current frame, we introduce simple yet effective memory selection strategies to provide more relevant information to the decoders, significantly improving performance. Furthermore, instead of parallel spatial and temporal localization, we propose a cascaded spatio-temporal design that connects the spatial decoder to the temporal decoder, allowing fine-grained spatial cues to assist complex temporal localization in long videos. Experiments on newly extended LF-STVG datasets show that ART-STVG significantly outperforms state-of-the-art methods, while achieving competitive performance on conventional short-form STVG.

Jiali Yao, Xinran Deng, Xin Gu et al.

In this paper, we propose spatio-temporal omni-object video grounding, dubbed OmniSTVG, a new STVG task that aims at localizing spatially and temporally all targets mentioned in the textual query from videos. Compared to classic STVG locating only a single target, OmniSTVG enables localization of not only an arbitrary number of text-referred targets but also their interacting counterparts in the query from the video, making it more flexible and practical in real scenarios for comprehensive understanding. In order to facilitate exploration of OmniSTVG, we introduce BOSTVG, a large-scale benchmark dedicated to OmniSTVG. Specifically, our BOSTVG consists of 10,018 videos with 10.2M frames and covers a wide selection of 287 classes from diverse scenarios. Each sequence in BOSTVG, paired with a free-form textual query, encompasses a varying number of targets ranging from 1 to 10. To ensure high quality, each video is manually annotated with meticulous inspection and refinement. To our best knowledge, BOSTVG is to date the first and the largest benchmark for OmniSTVG. To encourage future research, we introduce a simple yet effective approach, named OmniTube, which, drawing inspiration from Transformer-based STVG methods, is specially designed for OmniSTVG and demonstrates promising results. By releasing BOSTVG, we hope to go beyond classic STVG by locating every object appearing in the query for more comprehensive understanding, opening up a new direction for STVG. Our benchmark, model, and results will be released at https://github.com/JellyYao3000/OmniSTVG.

Yingjian Zhu, Ying Wang, Yuyang Hong et al.

Recently, an audio-visual instance segmentation (AVIS) task has been introduced, aiming to identify, segment and track individual sounding instances in videos. However, prevailing methods primarily adopt the offline paradigm, that cannot associate detected instances across consecutive clips, making them unsuitable for real-world scenarios that involve continuous video streams. To address this limitation, we introduce SeaVIS, the first online framework designed for audio-visual instance segmentation. SeaVIS leverages the Causal Cross Attention Fusion (CCAF) module to enable efficient online processing, which integrates visual features from the current frame with the entire audio history under strict causal constraints. A major challenge for conventional VIS methods is that appearance-based instance association fails to distinguish between an object's sounding and silent states, resulting in the incorrect segmentation of silent objects. To tackle this, we employ an Audio-Guided Contrastive Learning (AGCL) strategy to generate instance prototypes that encode not only visual appearance but also sounding activity. In this way, instances preserved during per-frame prediction that do not emit sound can be effectively suppressed during instance association process, thereby significantly enhancing the audio-following capability of SeaVIS. Extensive experiments conducted on the AVISeg dataset demonstrate that SeaVIS surpasses existing state-of-the-art models across multiple evaluation metrics while maintaining a competitive inference speed suitable for real-time processing.

Vidi Team, Chia-Wen Kuo, Chuang Huang et al.

Video has emerged as the primary medium for communication and creativity on the Internet, driving strong demand for scalable, high-quality video production. Vidi models continue to evolve toward next-generation video creation and have achieved state-of-the-art performance in multimodal temporal retrieval (TR). In its second release, Vidi2 advances video understanding with fine-grained spatio-temporal grounding (STG) and extends its capability to video question answering (Video QA), enabling comprehensive multimodal reasoning. Given a text query, Vidi2 can identify not only the corresponding timestamps but also the bounding boxes of target objects within the output time ranges. To enable comprehensive evaluation of STG, we introduce a new benchmark, VUE-STG, which offers critical improvements over existing STG datasets. In addition, we upgrade the previous VUE-TR benchmark to VUE-TR-V2, achieving a more balanced duration and query distribution. Remarkably, the Vidi2 model substantially outperforms leading proprietary systems, such as Gemini 3 Pro Preview and GPT-5, on both VUE-TR-V2 and VUE-STG, while achieving competitive results with popular open-source models with similar scale on video QA benchmarks. The latest Vidi2.5 offers significantly stronger STG capability and slightly better TR and Video QA performance over Vidi2. This update also introduces a Vidi2.5-Think model to handle plot understanding with complex plot reasoning. To comprehensively evaluate the performance of plot understanding, we propose VUE-PLOT benchmark with two tracks, Character and Reasoning. Notably, Vidi2.5-Think outperforms Gemini 3 Pro Preview on fine-grained character understanding with comparable performance on complex plot reasoning. Furthermore, we demonstrate the effectiveness of Vidi2.5 on a challenging real-world application, video editing planning.

Yijun Hu, Bing Fan, Xin Gu et al.

Establishing object-level correspondence between egocentric and exocentric views is essential for intelligent assistants to deliver precise and intuitive visual guidance. However, this task faces numerous challenges, including extreme viewpoint variations, occlusions, and the presence of small objects. Existing approaches usually borrow solutions from video object segmentation models, but still suffer from the aforementioned challenges. Recently, the Segment Anything Model 2 (SAM 2) has shown strong generalization capabilities and excellent performance in video object segmentation. Yet, when simply applied to the ego-exo correspondence (EEC) task, SAM 2 encounters severe difficulties due to ineffective ego-exo feature fusion and limited long-term memory capacity, especially for long videos. Addressing these problems, we propose a novel EEC framework based on SAM 2 with long-term memories by presenting a dual-memory architecture and an adaptive feature routing module inspired by Mixture-of-Experts (MoE). Compared to SAM 2, our approach features (i) a Memory-View MoE module which consists of a dual-branch routing mechanism to adaptively assign contribution weights to each expert feature along both channel and spatial dimensions, and (ii) a dual-memory bank system with a simple yet effective compression strategy to retain critical long-term information while eliminating redundancy. In the extensive experiments on the challenging EgoExo4D benchmark, our method, dubbed LM-EEC, achieves new state-of-the-art results and significantly outperforms existing methods and the SAM 2 baseline, showcasing its strong generalization across diverse scenarios. Our code and model are available at https://github.com/juneyeeHu/LM-EEC.

Ming Li, Xin Gu, Fan Chen et al.

Due to the challenges of manually collecting accurate editing data, existing datasets are typically constructed using various automated methods, leading to noisy supervision signals caused by the mismatch between editing instructions and original-edited image pairs. Recent efforts attempt to improve editing models through generating higher-quality edited images, pre-training on recognition tasks, or introducing vision-language models (VLMs) but fail to resolve this fundamental issue. In this paper, we offer a novel solution by constructing more effective editing instructions for given image pairs. This includes rectifying the editing instructions to better align with the original-edited image pairs and using contrastive editing instructions to further enhance their effectiveness. Specifically, we find that editing models exhibit specific generation attributes at different inference steps, independent of the text. Based on these prior attributes, we define a unified guide for VLMs to rectify editing instructions. However, there are some challenging editing scenarios that cannot be resolved solely with rectified instructions. To this end, we further construct contrastive supervision signals with positive and negative instructions and introduce them into the model training using triplet loss, thereby further facilitating supervision effectiveness. Our method does not require the VLM modules or pre-training tasks used in previous work, offering a more direct and efficient way to provide better supervision signals, and providing a novel, simple, and effective solution for instruction-based image editing. Results on multiple benchmarks demonstrate that our method significantly outperforms existing approaches. Compared with previous SOTA SmartEdit, we achieve 9.19% improvements on the Real-Edit benchmark with 30x less training data and 13x smaller model size.

Ruiheng Liu, Haihong Hao, Mingfei Han et al.

Advancing towards artificial superintelligence requires rich and intelligent perceptual capabilities. A critical frontier in this pursuit is overcoming the limited spatial understanding of Multimodal Large Language Models (MLLMs), where geometry information is essential. Existing methods often address this by rigidly injecting geometric signals into every input, while ignoring their necessity and adding computation overhead. Contrary to this paradigm, our framework endows the model with an awareness of perceptual insufficiency, empowering it to autonomously engage geometric features in reasoning when 2D cues are deemed insufficient. To achieve this, we first introduce an independent geometry input channel to the model architecture and conduct alignment training, enabling the effective utilization of geometric features. Subsequently, to endow the model with perceptual awareness, we curate a dedicated spatial-aware supervised fine-tuning dataset. This serves to activate the model's latent internal cues, empowering it to autonomously determine the necessity of geometric information. Experiments across multiple spatial reasoning benchmarks validate this approach, demonstrating significant spatial gains without compromising 2D visual reasoning capabilities, offering a path toward more robust, efficient and self-aware multi-modal intelligence.

Xin Gu, Yaojie Shen, Chenxi Luo et al.

Transformer has attracted increasing interest in STVG, owing to its end-to-end pipeline and promising result. Existing Transformer-based STVG approaches often leverage a set of object queries, which are initialized simply using zeros and then gradually learn target position information via iterative interactions with multimodal features, for spatial and temporal localization. Despite simplicity, these zero object queries, due to lacking target-specific cues, are hard to learn discriminative target information from interactions with multimodal features in complicated scenarios (\e.g., with distractors or occlusion), resulting in degradation. Addressing this, we introduce a novel Target-Aware Transformer for STVG (TA-STVG), which seeks to adaptively generate object queries via exploring target-specific cues from the given video-text pair, for improving STVG. The key lies in two simple yet effective modules, comprising text-guided temporal sampling (TTS) and attribute-aware spatial activation (ASA), working in a cascade. The former focuses on selecting target-relevant temporal cues from a video utilizing holistic text information, while the latter aims at further exploiting the fine-grained visual attribute information of the object from previous target-aware temporal cues, which is applied for object query initialization. Compared to existing methods leveraging zero-initialized queries, object queries in our TA-STVG, directly generated from a given video-text pair, naturally carry target-specific cues, making them adaptive and better interact with multimodal features for learning more discriminative information to improve STVG. In our experiments on three benchmarks, TA-STVG achieves state-of-the-art performance and significantly outperforms the baseline, validating its efficacy.

Guanlin He, Yingtai Xiao, Jiamu Bai et al.

Matrix mechanisms are often used to provide unbiased differentially private query answers when publishing statistics or creating synthetic data. Recent work has developed matrix mechanisms, such as ResidualPlanner and Weighted Fourier Factorizations, that scale to high dimensional datasets while providing optimality guarantees for workloads such as marginals and circular product queries. They operate by adding noise to a linearly independent set of queries that can compactly represent the desired workloads. In this paper, we present QuerySmasher, an alternative scalable approach based on a divide-and-conquer strategy. Given a workload that can be answered from various data marginals, QuerySmasher splits each query into sub-queries and re-assembles the pieces into mutually orthogonal sub-workloads. These sub-workloads represent small, low-dimensional problems that can be independently and optimally answered by existing low-dimensional matrix mechanisms. QuerySmasher then stitches these solutions together to answer queries in the original workload. We show that QuerySmasher subsumes prior work, like ResidualPlanner (RP), ResidualPlanner+ (RP+), and Weighted Fourier Factorizations (WFF). We prove that it can dominate those approaches, under sum squared error, for all workloads. We also experimentally demonstrate the scalability and accuracy of QuerySmasher.

Vidi Team, Celong Liu, Chia-Wen Kuo et al.

Humans naturally share information with those they are connected to, and video has become one of the dominant mediums for communication and expression on the Internet. To support the creation of high-quality large-scale video content, a modern pipeline requires a comprehensive understanding of both the raw input materials (e.g., the unedited footage captured by cameras) and the editing components (e.g., visual effects). In video editing scenarios, models must process multiple modalities (e.g., vision, audio, text) with strong background knowledge and handle flexible input lengths (e.g., hour-long raw videos), which poses significant challenges for traditional models. In this report, we introduce Vidi, a family of Large Multimodal Models (LMMs) for a wide range of video understand editing scenarios. The first release focuses on temporal retrieval, i.e., identifying the time ranges within the input videos corresponding to a given text query, which plays a critical role in intelligent editing. The model is capable of processing hour-long videos with strong temporal understanding capability, e.g., retrieve time ranges for certain queries. To support a comprehensive evaluation in real-world scenarios, we also present the VUE-TR benchmark, which introduces five key advancements. 1) Video duration: significantly longer than videos of existing temporal retrival datasets, 2) Audio support: includes audio-based queries, 3) Query format: diverse query lengths/formats, 4) Annotation quality: ground-truth time ranges are manually annotated. 5) Evaluation metric: a refined IoU metric to support evaluation over multiple time ranges. Remarkably, Vidi significantly outperforms leading proprietary models, e.g., GPT-4o and Gemini, on the temporal retrieval task, indicating its superiority in video editing scenarios.

Donghwan Kim, Xin Gu, Jinho Baek et al.

Machine learning (ML) models memorize and leak training data, causing serious privacy issues to data owners. Training algorithms with differential privacy (DP), such as DP-SGD, have been gaining attention as a solution. However, DP-SGD adds a noise at each training iteration, which degrades the accuracy of the trained model. To improve accuracy, a new family of approaches adds carefully designed correlated noises, so that noises cancel out each other across iterations. We performed an extensive characterization study of these new mechanisms, for the first time to the best of our knowledge, and show they incur non-negligible overheads when the model is large or uses large embedding tables. Motivated by the analysis, we propose Cocoon, a hardware-software co-designed framework for efficient training with correlated noises. Cocoon accelerates models with embedding tables through pre-computing and storing correlated noises in a coalesced format (Cocoon-Emb), and supports large models through a custom near-memory processing device (Cocoon-NMP). On a real system with an FPGA-based NMP device prototype, Cocoon improves the performance by 2.33-10.82x(Cocoon-Emb) and 1.55-3.06x (Cocoon-NMP).

Xin Gu, Yingtai Xiao, Guanlin He et al.

Differentially private stochastic gradient descent (DP-SGD) offers the promise of training deep learning models while mitigating many privacy risks. However, there is currently a large accuracy gap between DP-SGD and normal SGD training. This has resulted in different lines of research investigating orthogonal ways of improving privacy-preserving training. One such line of work, known as DP-MF, correlates the privacy noise across different iterations of stochastic gradient descent -- allowing later iterations to cancel out some of the noise added to earlier iterations. In this paper, we study how to improve this noise correlation. We propose a technique called NoiseCurve that uses model curvature, estimated from public unlabeled data, to improve the quality of this cross-iteration noise correlation. Our experiments on various datasets, models, and privacy parameters show that the noise correlations computed by NoiseCurve offer consistent and significant improvements in accuracy over the correlation scheme used by DP-MF.