Long video understanding poses unique challenges due to their temporal complexity and low information density. Recent works address this task by sampling numerous frames or incorporating auxiliary tools using LLMs, both of which result in high computational costs. In this work, we introduce a curiosity-driven video agent with self-exploration capability, dubbed as VCA. Built upon VLMs, VCA autonomously navigates video segments and efficiently builds a comprehensive understanding of complex video sequences. Instead of directly sampling frames, VCA employs a tree-search structure to explore video segments and collect frames. Rather than relying on external feedback or reward, VCA leverages VLM's self-generated intrinsic reward to guide its exploration, enabling it to capture the most crucial information for reasoning. Experimental results on multiple long video benchmarks demonstrate our approach's superior effectiveness and efficiency.

Large language models (LLMs) have demonstrated remarkable capabilities across a range of text-generation tasks. However, LLMs still struggle with problems requiring multi-step decision-making and environmental feedback, such as online shopping, scientific reasoning, and mathematical problem-solving. Unlike pure text data, collecting large-scale decision-making data is challenging. Moreover, many powerful LLMs are only accessible through APIs, which hinders their fine-tuning for agent tasks due to cost and complexity. To address LLM agents' limitations, we propose a framework that can automatically learn a reward model from the environment without human annotations. This model can be used to evaluate the action trajectories of LLM agents and provide heuristics for task planning. Specifically, our approach involves employing one LLM-based agent to navigate an environment randomly, generating diverse action trajectories. Subsequently, a separate LLM is leveraged to assign a task intent and synthesize a negative response alongside the correct response for each trajectory. These triplets (task intent, positive response, and negative response) are then utilized as training data to optimize a reward model capable of scoring action trajectories. This reward model can be integrated with LLM-based agents and various planning algorithms to enhance task-solving performance. The effectiveness and generalizability of our framework are demonstrated through evaluations conducted on different agent benchmarks. In conclusion, our proposed framework represents a significant advancement in enhancing LLM agents' decision-making capabilities. By automating the learning of reward models, we overcome the challenges of data scarcity and API limitations, potentially revolutionizing the application of LLMs in complex and interactive environments. This research paves the way for more sophisticated AI agents capable of tackling a wide range of real-world problems requiring multi-step decision-making.

Longxuan Yu, Delin Chen, Siheng Xiong, Qingyang Wu, Dawei Li, Zhikai Chen, Xiaoze Liu, Liangming Pan. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

A remarkable ability of human beings resides in compositional reasoning, i.e., the capacity to make "infinite use of finite means". However, current large vision-language foundation models (VLMs) fall short of such compositional abilities due to their ``bag-of-words" behaviors and inability to construct words that correctly represent visual entities and the relations among the entities. To this end, we propose CoVLM, which can guide the LLM to explicitly compose visual entities and relationships among the text and dynamically communicate with the vision encoder and detection network to achieve vision-language communicative decoding. Specifically, we first devise a set of novel communication tokens for the LLM, for dynamic communication between the visual detection system and the language system. A communication token is generated by the LLM following a visual entity or a relation, to inform the detection network to propose regions that are relevant to the sentence generated so far. The proposed regions-of-interests (ROIs) are then fed back into the LLM for better language generation contingent on the relevant regions. The LLM is thus able to compose the visual entities and relationships through the communication tokens. The vision-to-language and language-to-vision communication are iteratively performed until the entire sentence is generated. Our framework seamlessly bridges the gap between visual perception and LLMs and outperforms previous VLMs by a large margin on compositional reasoning benchmarks (e.g., ~20% in HICO-DET mAP, ~14% in Cola top-1 accuracy, and ~3% on ARO top-1 accuracy). We also achieve state-of-the-art performances on traditional vision-language tasks such as referring expression comprehension and visual question answering.

Although Magnetic Resonance Imaging (MRI) is very helpful for brain tumor segmentation and discovery, it often lacks some modalities in clinical practice. As a result, degradation of prediction performance is inevitable. According to current implementations, different modalities are considered to be independent and non-interfering with each other during the training process of modal feature extraction, however they are complementary. In this paper, considering the sensitivity of different modalities to diverse tumor regions, we propose a Category Aware Group Self-Support Learning framework, called GSS, to make up for the information deficit among the modalities in the individual modal feature extraction phase. Precisely, within each prediction category, predictions of all modalities form a group, where the prediction with the most extraordinary sensitivity is selected as the group leader. Collaborative efforts between group leaders and members identify the communal learning target with high consistency and certainty. As our minor contribution, we introduce a random mask to reduce the possible biases. GSS adopts the standard training strategy without specific architectural choices and thus can be easily plugged into existing incomplete multi-modal brain tumor segmentation. Remarkably, extensive experiments on BraTS2020, BraTS2018, and BraTS2015 datasets demonstrate that GSS can improve the performance of existing SOTA algorithms by 1.27-3.20% in Dice on average. The code is released at https://github.com/qysgithubopen/GSS.

Cluster motors are widely used in daily production. Both simple manual inspection and invasive monitoring of cluster motors have the drawbacks of high workload, high cost, and low efficiency. This paper presents a method of multi-state non-invasive monitoring of cluster motors by collecting single motor data with three-phase electrical instruments. The data collected by monitoring is transmitted to the environmental monitoring host through LoRa wireless transmission, and then identified by computer programming. Based on Grubbs criteria, the state of each motor in the cluster motors is analyzed, its working state is determined, and its single motor startup, shutdown and working condition switching are identified, which is helpful to the management of the cluster motors.

In the realm of medical imaging, distinct magnetic resonance imaging (MRI) modalities can provide complementary medical insights. However, it is not uncommon for one or more modalities to be absent due to image corruption, artifacts, acquisition protocols, allergies to contrast agents, or cost constraints, posing a significant challenge for perceiving the modality-absent state in incomplete modality segmentation.In this work, we introduce a novel incomplete multi-modal segmentation framework called Modal-aware Visual Prompting (MAVP), which draws inspiration from the widely used pre-training and prompt adjustment protocol employed in natural language processing (NLP). In contrast to previous prompts that typically use textual network embeddings, we utilize embeddings as the prompts generated by a modality state classifier that focuses on the missing modality states. Additionally, we integrate modality state prompts into both the extraction stage of each modality and the modality fusion stage to facilitate intra/inter-modal adaptation. Our approach achieves state-of-the-art performance in various modality-incomplete scenarios compared to incomplete modality-specific solutions.

Line detection is an important research topic in image processing and computer vision. Hough transform is a widely used technique to detect lines. In the scenario of cloud computing, performing the Hough transform and line detection needs to consider privacy protection issues. In this paper, we propose implementing Hough transform in the encrypted domain and its application to line detection on encrypted images. Using the proposed encrypted domain Hough transform, we detect the extreme points in the encrypted parameter space. After transforming the extreme point into the encrypted spatial domain, we obtain the detected lines on the encrypted image. We conducted experiments and demonstrated the viability and effectiveness of our secure Hough transform and line detection on encrypted images.

Edge detection is one of the most important topics of image processing. In the scenario of cloud computing, performing edge detection may also consider privacy protection. In this paper, we propose an edge detection and image segmentation scheme on an encrypted image with Sobel edge detector. We implement Gaussian filtering and Sobel operator on the image in the encrypted domain with homomorphic property. By implementing an adaptive threshold decision algorithm in the encrypted domain, we obtain a threshold determined by the image distribution. With the technique of garbled circuit, we perform comparison in the encrypted domain and obtain the edge of the image without decrypting the image in advanced. We then propose an image segmentation scheme on the encrypted image based on the detected edges. Our experiments demonstrate the viability and effectiveness of the proposed encrypted image edge detection and segmentation.