LLaVA-MLB: Mitigating and Leveraging Attention Bias for Training-Free Video LLMs

Training-free video large language models (LLMs) leverage pretrained Image LLMs to process video content without the need for further training. A key challenge in such approaches is the difficulty of retaining essential visual and temporal information, constrained by the token limits in Image LLMs. To address this, we propose a two-stage method for selecting query-relevant tokens based on the LLM attention scores: compressing the video sequence and then expanding the sequence. However, during the compression stage, Image LLMs often exhibit a positional attention bias in video sequences, where attention is overly concentrated on later frames, causing early-frame information to be underutilized. To alleviate this attention bias during sequence compression, we propose Gridded Attention Pooling for preserving spatiotemporal structure. Additionally, we introduce Visual Summarization Tail to effectively utilize this bias, facilitating overall video understanding during sequence expansion. In this way, our method effectively Mitigates and Leverages attention Bias (LLaVA-MLB), enabling the frozen Image LLM for detailed video understanding. Experiments on several benchmarks demonstrate that our approach outperforms state-of-the-art methods, achieving superior performance in both efficiency and accuracy. Our code will be released.

FastVID: Dynamic Density Pruning for Fast Video Large Language Models

Video Large Language Models have shown impressive capabilities in video comprehension, yet their practical deployment is hindered by substantial inference costs caused by redundant video tokens. Existing pruning techniques fail to fully exploit the spatiotemporal redundancy inherent in video data. To bridge this gap, we perform a systematic analysis of video redundancy from two perspectives: temporal context and visual context. Leveraging this insight, we propose Dynamic Density Pruning for Fast Video LLMs termed FastVID. Specifically, FastVID dynamically partitions videos into temporally ordered segments to preserve temporal structure and applies a density-based token pruning strategy to maintain essential visual information. Our method significantly reduces computational overhead while maintaining temporal and visual integrity. Extensive evaluations show that FastVID achieves state-of-the-art performance across various short- and long-video benchmarks on leading Video LLMs, including LLaVA-OneVision and LLaVA-Video. Notably, FastVID effectively prunes 90% of video tokens while retaining 98.0% of LLaVA-OneVision's original performance. The code is available at https://github.com/LunarShen/FastVID.

You Only Click Once: Single Point Weakly Supervised 3D Instance Segmentation for Autonomous Driving

Outdoor LiDAR point cloud 3D instance segmentation is a crucial task in autonomous driving. However, it requires laborious human efforts to annotate the point cloud for training a segmentation model. To address this challenge, we propose a YoCo framework, which generates 3D pseudo labels using minimal coarse click annotations in the bird's eye view plane. It is a significant challenge to produce high-quality pseudo labels from sparse annotations. Our YoCo framework first leverages vision foundation models combined with geometric constraints from point clouds to enhance pseudo label generation. Second, a temporal and spatial-based label updating module is designed to generate reliable updated labels. It leverages predictions from adjacent frames and utilizes the inherent density variation of point clouds (dense near, sparse far). Finally, to further improve label quality, an IoU-guided enhancement module is proposed, replacing pseudo labels with high-confidence and high-IoU predictions. Experiments on the Waymo dataset demonstrate YoCo's effectiveness and generality, achieving state-of-the-art performance among weakly supervised methods and surpassing fully supervised Cylinder3D. Additionally, the YoCo is suitable for various networks, achieving performance comparable to fully supervised methods with minimal fine-tuning using only 0.8% of the fully labeled data, significantly reducing annotation costs.

Qwen2.5-1M Technical Report

We introduce Qwen2.5-1M, a series of models that extend the context length to 1 million tokens. Compared to the previous 128K version, the Qwen2.5-1M series have significantly enhanced long-context capabilities through long-context pre-training and post-training. Key techniques such as long data synthesis, progressive pre-training, and multi-stage supervised fine-tuning are employed to effectively enhance long-context performance while reducing training costs. To promote the use of long-context models among a broader user base, we present and open-source our inference framework. This framework includes a length extrapolation method that can expand the model context lengths by at least four times, or even more, without additional training. To reduce inference costs, we implement a sparse attention method along with chunked prefill optimization for deployment scenarios and a sparsity refinement method to improve precision. Additionally, we detail our optimizations in the inference engine, including kernel optimization, pipeline parallelism, and scheduling optimization, which significantly enhance overall inference performance. By leveraging our inference framework, the Qwen2.5-1M models achieve a remarkable 3x to 7x prefill speedup in scenarios with 1 million tokens of context. This framework provides an efficient and powerful solution for developing applications that require long-context processing using open-source models. The Qwen2.5-1M series currently includes the open-source models Qwen2.5-7B-Instruct-1M and Qwen2.5-14B-Instruct-1M, as well as the API-accessed model Qwen2.5-Turbo. Evaluations show that Qwen2.5-1M models have been greatly improved in long-context tasks without compromising performance in short-context scenarios. Specifically, the Qwen2.5-14B-Instruct-1M model significantly outperforms GPT-4o-mini in long-context tasks and supports contexts eight times longer.

Generative Planning with 3D-vision Language Pre-training for End-to-End Autonomous Driving

Autonomous driving is a challenging task that requires perceiving and understanding the surrounding environment for safe trajectory planning. While existing vision-based end-to-end models have achieved promising results, these methods are still facing the challenges of vision understanding, decision reasoning and scene generalization. To solve these issues, a generative planning with 3D-vision language pre-training model named GPVL is proposed for end-to-end autonomous driving. The proposed paradigm has two significant aspects. On one hand, a 3D-vision language pre-training module is designed to bridge the gap between visual perception and linguistic understanding in the bird's eye view. On the other hand, a cross-modal language model is introduced to generate holistic driving decisions and fine-grained trajectories with perception and navigation information in an auto-regressive manner. Experiments on the challenging nuScenes dataset demonstrate that the proposed scheme achieves excellent performances compared with state-of-the-art methods. Besides, the proposed GPVL presents strong generalization ability and real-time potential when handling high-level commands in various scenarios. It is believed that the effective, robust and efficient performance of GPVL is crucial for the practical application of future autonomous driving systems. Code is available at https://github.com/ltp1995/GPVL

Simulation-Free Hierarchical Latent Policy Planning for Proactive Dialogues

Recent advancements in proactive dialogues have garnered significant attention, particularly for more complex objectives (e.g. emotion support and persuasion). Unlike traditional task-oriented dialogues, proactive dialogues demand advanced policy planning and adaptability, requiring rich scenarios and comprehensive policy repositories to develop such systems. However, existing approaches tend to rely on Large Language Models (LLMs) for user simulation and online learning, leading to biases that diverge from realistic scenarios and result in suboptimal efficiency. Moreover, these methods depend on manually defined, context-independent, coarse-grained policies, which not only incur high expert costs but also raise concerns regarding their completeness. In our work, we highlight the potential for automatically discovering policies directly from raw, real-world dialogue records. To this end, we introduce a novel dialogue policy planning framework, LDPP. It fully automates the process from mining policies in dialogue records to learning policy planning. Specifically, we employ a variant of the Variational Autoencoder to discover fine-grained policies represented as latent vectors. After automatically annotating the data with these latent policy labels, we propose an Offline Hierarchical Reinforcement Learning (RL) algorithm in the latent space to develop effective policy planning capabilities. Our experiments demonstrate that LDPP outperforms existing methods on two proactive scenarios, even surpassing ChatGPT with only a 1.8-billion-parameter LLM.

A Lightweight U-like Network Utilizing Neural Memory Ordinary Differential Equations for Slimming the Decoder

In recent years, advanced U-like networks have demonstrated remarkable performance in medical image segmentation tasks. However, their drawbacks, including excessive parameters, high computational complexity, and slow inference speed, pose challenges for practical implementation in scenarios with limited computational resources. Existing lightweight U-like networks have alleviated some of these problems, but they often have pre-designed structures and consist of inseparable modules, limiting their application scenarios. In this paper, we propose three plug-and-play decoders by employing different discretization methods of the neural memory Ordinary Differential Equations (nmODEs). These decoders integrate features at various levels of abstraction by processing information from skip connections and performing numerical operations on upward path. Through experiments on the PH2, ISIC2017, and ISIC2018 datasets, we embed these decoders into different U-like networks, demonstrating their effectiveness in significantly reducing the number of parameters and FLOPs while maintaining performance. In summary, the proposed discretized nmODEs decoders are capable of reducing the number of parameters by about 20% ~ 50% and FLOPs by up to 74%, while possessing the potential to adapt to all U-like networks. Our code is available at https://github.com/nayutayuki/Lightweight-nmODE-Decoders-For-U-like-networks.

Towards Effective Data-Free Knowledge Distillation via Diverse Diffusion Augmentation

Data-free knowledge distillation (DFKD) has emerged as a pivotal technique in the domain of model compression, substantially reducing the dependency on the original training data. Nonetheless, conventional DFKD methods that employ synthesized training data are prone to the limitations of inadequate diversity and discrepancies in distribution between the synthesized and original datasets. To address these challenges, this paper introduces an innovative approach to DFKD through diverse diffusion augmentation (DDA). Specifically, we revise the paradigm of common data synthesis in DFKD to a composite process through leveraging diffusion models subsequent to data synthesis for self-supervised augmentation, which generates a spectrum of data samples with similar distributions while retaining controlled variations. Furthermore, to mitigate excessive deviation in the embedding space, we introduce an image filtering technique grounded in cosine similarity to maintain fidelity during the knowledge distillation process. Comprehensive experiments conducted on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets showcase the superior performance of our method across various teacher-student network configurations, outperforming the contemporary state-of-the-art DFKD methods. Code will be available at:https://github.com/SLGSP/DDA.

MuAP: Multi-step Adaptive Prompt Learning for Vision-Language Model with Missing Modality

Recently, prompt learning has garnered considerable attention for its success in various Vision-Language (VL) tasks. However, existing prompt-based models are primarily focused on studying prompt generation and prompt strategies with complete modality settings, which does not accurately reflect real-world scenarios where partial modality information may be missing. In this paper, we present the first comprehensive investigation into prompt learning behavior when modalities are incomplete, revealing the high sensitivity of prompt-based models to missing modalities. To this end, we propose a novel Multi-step Adaptive Prompt Learning (MuAP) framework, aiming to generate multimodal prompts and perform multi-step prompt tuning, which adaptively learns knowledge by iteratively aligning modalities. Specifically, we generate multimodal prompts for each modality and devise prompt strategies to integrate them into the Transformer model. Subsequently, we sequentially perform prompt tuning from single-stage and alignment-stage, allowing each modality-prompt to be autonomously and adaptively learned, thereby mitigating the imbalance issue caused by only textual prompts that are learnable in previous works. Extensive experiments demonstrate the effectiveness of our MuAP and this model achieves significant improvements compared to the state-of-the-art on all benchmark datasets

Strengthening Layer Interaction via Dynamic Layer Attention

In recent years, employing layer attention to enhance interaction among hierarchical layers has proven to be a significant advancement in building network structures. In this paper, we delve into the distinction between layer attention and the general attention mechanism, noting that existing layer attention methods achieve layer interaction on fixed feature maps in a static manner. These static layer attention methods limit the ability for context feature extraction among layers. To restore the dynamic context representation capability of the attention mechanism, we propose a Dynamic Layer Attention (DLA) architecture. The DLA comprises dual paths, where the forward path utilizes an improved recurrent neural network block, named Dynamic Sharing Unit (DSU), for context feature extraction. The backward path updates features using these shared context representations. Finally, the attention mechanism is applied to these dynamically refreshed feature maps among layers. Experimental results demonstrate the effectiveness of the proposed DLA architecture, outperforming other state-of-the-art methods in image recognition and object detection tasks. Additionally, the DSU block has been evaluated as an efficient plugin in the proposed DLA architecture.The code is available at https://github.com/tunantu/Dynamic-Layer-Attention.