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.

Bridge then Begin Anew: Generating Target-relevant Intermediate Model for Source-free Visual Emotion Adaptation

Visual emotion recognition (VER), which aims at understanding humans' emotional reactions toward different visual stimuli, has attracted increasing attention. Given the subjective and ambiguous characteristics of emotion, annotating a reliable large-scale dataset is hard. For reducing reliance on data labeling, domain adaptation offers an alternative solution by adapting models trained on labeled source data to unlabeled target data. Conventional domain adaptation methods require access to source data. However, due to privacy concerns, source emotional data may be inaccessible. To address this issue, we propose an unexplored task: source-free domain adaptation (SFDA) for VER, which does not have access to source data during the adaptation process. To achieve this, we propose a novel framework termed Bridge then Begin Anew (BBA), which consists of two steps: domain-bridged model generation (DMG) and target-related model adaptation (TMA). First, the DMG bridges cross-domain gaps by generating an intermediate model, avoiding direct alignment between two VER datasets with significant differences. Then, the TMA begins training the target model anew to fit the target structure, avoiding the influence of source-specific knowledge. Extensive experiments are conducted on six SFDA settings for VER. The results demonstrate the effectiveness of BBA, which achieves remarkable performance gains compared with state-of-the-art SFDA methods and outperforms representative unsupervised domain adaptation approaches.

From Easy to Hard: Progressive Active Learning Framework for Infrared Small Target Detection with Single Point Supervision

Recently, single-frame infrared small target (SIRST) detection with single point supervision has drawn wide-spread attention. However, the latest label evolution with single point supervision (LESPS) framework suffers from instability, excessive label evolution, and difficulty in exerting embedded network performance. Therefore, we construct a Progressive Active Learning (PAL) framework. Specifically, inspired by organisms gradually adapting to their environment and continuously accumulating knowledge, we propose an innovative progressive active learning idea, which emphasizes that the network progressively and actively recognizes and learns more hard samples to achieve continuous performance enhancement. Based on this, on the one hand, we propose a model pre-start concept, which focuses on selecting a portion of easy samples and can help models have basic task-specific learning capabilities. On the other hand, we propose a refined dual-update strategy, which can promote reasonable learning of harder samples and continuous refinement of pseudo-labels. In addition, to alleviate the risk of excessive label evolution, a decay factor is reasonably introduced, which helps to achieve a dynamic balance between the expansion and contraction of target annotations. Extensive experiments show that convolutional neural networks (CNNs) equipped with our PAL framework have achieved state-of-the-art (SOTA) results on multiple public datasets. Furthermore, our PAL framework can build a efficient and stable bridge between full supervision and point supervision tasks. Our code are available at https://github.com/YuChuang1205/PAL.

HEIE: MLLM-Based Hierarchical Explainable AIGC Image Implausibility Evaluator

AIGC images are prevalent across various fields, yet they frequently suffer from quality issues like artifacts and unnatural textures. Specialized models aim to predict defect region heatmaps but face two primary challenges: (1) lack of explainability, failing to provide reasons and analyses for subtle defects, and (2) inability to leverage common sense and logical reasoning, leading to poor generalization. Multimodal large language models (MLLMs) promise better comprehension and reasoning but face their own challenges: (1) difficulty in fine-grained defect localization due to the limitations in capturing tiny details; and (2) constraints in providing pixel-wise outputs necessary for precise heatmap generation. To address these challenges, we propose HEIE: a novel MLLM-Based Hierarchical Explainable image Implausibility Evaluator. We introduce the CoT-Driven Explainable Trinity Evaluator, which integrates heatmaps, scores, and explanation outputs, using CoT to decompose complex tasks into subtasks of increasing difficulty and enhance interpretability. Our Adaptive Hierarchical Implausibility Mapper synergizes low-level image features with high-level mapper tokens from LLMs, enabling precise local-to-global hierarchical heatmap predictions through an uncertainty-based adaptive token approach. Moreover, we propose a new dataset: Expl-AIGI-Eval, designed to facilitate interpretable implausibility evaluation of AIGC images. Our method demonstrates state-of-the-art performance through extensive experiments.

TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval

Most text-video retrieval methods utilize the text-image pre-trained CLIP as a backbone, incorporating complex modules that result in high computational overhead. As a result, many studies focus on efficient fine-tuning. The primary challenge in efficient adaption arises from the inherent differences between image and video modalities. Each sampled video frame must be processed by the image encoder independently, which increases complexity and complicates practical deployment. Although existing efficient methods fine-tune with small trainable parameters, they still incur high inference costs due to the large token number. In this work, we argue that temporal redundancy significantly contributes to the model's high complexity due to the repeated information in consecutive frames. Existing token compression methods for image models fail to solve the unique challenges, as they overlook temporal redundancy across frames. To tackle these problems, we propose Temporal Token Merging (TempMe) to reduce temporal redundancy. Specifically, we introduce a progressive multi-granularity framework. By gradually combining neighboring clips, we merge temporal tokens across different frames and learn video-level features, leading to lower complexity and better performance. Extensive experiments validate the superiority of our TempMe. Compared to previous efficient text-video retrieval methods, TempMe significantly reduces output tokens by 95% and GFLOPs by 51%, while achieving a 1.8X speedup and a 4.4% R-Sum improvement. Additionally, TempMe exhibits robust generalization capabilities by integrating effectively with both efficient and full fine-tuning methods. With full fine-tuning, TempMe achieves a significant 7.9% R-Sum improvement, trains 1.57X faster, and utilizes 75.2% GPU memory usage. Our code will be released.

Multi-source Domain Adaptation for Panoramic Semantic Segmentation

Panoramic semantic segmentation has received widespread attention recently due to its comprehensive 360\degree field of view. However, labeling such images demands greater resources compared to pinhole images. As a result, many unsupervised domain adaptation methods for panoramic semantic segmentation have emerged, utilizing real pinhole images or low-cost synthetic panoramic images. But, the segmentation model lacks understanding of the panoramic structure when only utilizing real pinhole images, and it lacks perception of real-world scenes when only adopting synthetic panoramic images. Therefore, in this paper, we propose a new task of multi-source domain adaptation for panoramic semantic segmentation, aiming to utilize both real pinhole and synthetic panoramic images in the source domains, enabling the segmentation model to perform well on unlabeled real panoramic images in the target domain. Further, we propose Deformation Transform Aligner for Panoramic Semantic Segmentation (DTA4PASS), which converts all pinhole images in the source domains into panoramic-like images, and then aligns the converted source domains with the target domain. Specifically, DTA4PASS consists of two main components: Unpaired Semantic Morphing (USM) and Distortion Gating Alignment (DGA). Firstly, in USM, the Semantic Dual-view Discriminator (SDD) assists in training the diffeomorphic deformation network, enabling the effective transformation of pinhole images without paired panoramic views. Secondly, DGA assigns pinhole-like and panoramic-like features to each image by gating, and aligns these two features through uncertainty estimation. DTA4PASS outperforms the previous state-of-the-art methods by 1.92% and 2.19% on the outdoor and indoor multi-source domain adaptation scenarios, respectively. The source code will be released.

A Survey of Embodied Learning for Object-Centric Robotic Manipulation

Embodied learning for object-centric robotic manipulation is a rapidly developing and challenging area in embodied AI. It is crucial for advancing next-generation intelligent robots and has garnered significant interest recently. Unlike data-driven machine learning methods, embodied learning focuses on robot learning through physical interaction with the environment and perceptual feedback, making it especially suitable for robotic manipulation. In this paper, we provide a comprehensive survey of the latest advancements in this field and categorize the existing work into three main branches: 1) Embodied perceptual learning, which aims to predict object pose and affordance through various data representations; 2) Embodied policy learning, which focuses on generating optimal robotic decisions using methods such as reinforcement learning and imitation learning; 3) Embodied task-oriented learning, designed to optimize the robot's performance based on the characteristics of different tasks in object grasping and manipulation. In addition, we offer an overview and discussion of public datasets, evaluation metrics, representative applications, current challenges, and potential future research directions. A project associated with this survey has been established at https://github.com/RayYoh/OCRM_survey.

LLMI3D: Empowering LLM with 3D Perception from a Single 2D Image

Recent advancements in autonomous driving, augmented reality, robotics, and embodied intelligence have necessitated 3D perception algorithms. However, current 3D perception methods, particularly small models, struggle with processing logical reasoning, question-answering, and handling open scenario categories. On the other hand, generative multimodal large language models (MLLMs) excel in general capacity but underperform in 3D tasks, due to weak spatial and local object perception, poor text-based geometric numerical output, and inability to handle camera focal variations. To address these challenges, we propose the following solutions: Spatial-Enhanced Local Feature Mining for better spatial feature extraction, 3D Query Token-Derived Info Decoding for precise geometric regression, and Geometry Projection-Based 3D Reasoning for handling camera focal length variations. We employ parameter-efficient fine-tuning for a pre-trained MLLM and develop LLMI3D, a powerful 3D perception MLLM. Additionally, we have constructed the IG3D dataset, which provides fine-grained descriptions and question-answer annotations. Extensive experiments demonstrate that our LLMI3D achieves state-of-the-art performance, significantly outperforming existing methods.

Local Manifold Learning for No-Reference Image Quality Assessment

Contrastive learning has considerably advanced the field of Image Quality Assessment (IQA), emerging as a widely adopted technique. The core mechanism of contrastive learning involves minimizing the distance between quality-similar (positive) examples while maximizing the distance between quality-dissimilar (negative) examples. Despite its successes, current contrastive learning methods often neglect the importance of preserving the local manifold structure. This oversight can result in a high degree of similarity among hard examples within the feature space, thereby impeding effective differentiation and assessment. To address this issue, we propose an innovative framework that integrates local manifold learning with contrastive learning for No-Reference Image Quality Assessment (NR-IQA). Our method begins by sampling multiple crops from a given image, identifying the most visually salient crop. This crop is then used to cluster other crops from the same image as the positive class, while crops from different images are treated as negative classes to increase inter-class distance. Uniquely, our approach also considers non-saliency crops from the same image as intra-class negative classes to preserve their distinctiveness. Additionally, we employ a mutual learning framework, which further enhances the model's ability to adaptively learn and identify visual saliency regions. Our approach demonstrates a better performance compared to state-of-the-art methods in 7 standard datasets, achieving PLCC values of 0.942 (compared to 0.908 in TID2013) and 0.914 (compared to 0.894 in LIVEC).