Active Speaker Detection (ASD) aims to determine whether each candidate in a video frame is speaking. The egocentric dataset Ego4D introduces unique challenges for this task, such as dynamic shooting angles that cause candidates to frequently leave the sight, leading to temporal discontinuities. Additionally, Ego4D poses a novel task: detecting the speaking activities of the camera wearer, who never appears in the field of view. Existing methods treat these two tasks separately, and treat candidates out of sight as noise. In contrast, we propose EgoNet, a framework that uniformly models all candidates, including those not visible. By capturing interactions among all candidates and modeling broader temporal context, EgoNet reduces uncertainty and improves performance in egocentric active speaker detection.

We proposes a novel visual tokenizer by combining high-level semantic tokens and low-level pixel tokens to represent images, aiming to address the challenges of image-to-sequence conversion for Large Language Models (LLMs). Existing visual tokenizers, such as VQ-VAE and diffusion-based models, either struggle with token explosion as image resolution increases or fail to capture detailed structural information. Our method introduces a dual-token system: high-level semantic tokens capture the main content of the image, while low-level pixel tokens preserve structural details. By integrating these tokens in a hybrid architecture, we leverage a VQ-VAE branch to generate low-resolution guidance and a diffusion process to reconstruct high-resolution images with both semantic coherence and structural accuracy. This approach significantly reduces the number of required tokens and enhances image reconstruction quality, offering an efficient solution for tasks like image generation and understanding based on LLMs.

The development of reasoning capabilities represents a critical frontier in large language models (LLMs) research, where reinforcement learning (RL) and process reward models (PRMs) have emerged as predominant methodological frameworks. Contrary to conventional wisdom, empirical evidence from DeepSeek-R1 demonstrates that pure RL training focused on mathematical problem-solving can progressively enhance reasoning abilities without PRM integration, challenging the perceived necessity of process supervision. In this study, we conduct a systematic investigation of the relationship between RL training and PRM capabilities. Our findings demonstrate that problem-solving proficiency and process supervision capabilities represent complementary dimensions of reasoning that co-evolve synergistically during pure RL training. In particular, current PRMs underperform simple baselines like majority voting when applied to state-of-the-art models such as DeepSeek-R1 and QwQ-32B. To address this limitation, we propose Self-PRM, an introspective framework in which models autonomously evaluate and rerank their generated solutions through self-reward mechanisms. Although Self-PRM consistently improves the accuracy of the benchmark (particularly with larger sample sizes), analysis exposes persistent challenges: The approach exhibits low precision (<10\%) on difficult problems, frequently misclassifying flawed solutions as valid. These analyses underscore the need for combined training with process supervision and continued RL scaling to enhance reward alignment and introspective accuracy. We hope these findings provide actionable insights for building more reliable and self-aware complex reasoning models.

Detecting hand actions in videos is crucial for understanding video content and has diverse real-world applications. Existing approaches often focus on whole-body actions or coarse-grained action categories, lacking fine-grained hand-action localization information. To fill this gap, we introduce the FHA-Kitchens (Fine-Grained Hand Actions in Kitchen Scenes) dataset, providing both coarse- and fine-grained hand action categories along with localization annotations. This dataset comprises 2,377 video clips and 30,047 frames, annotated with approximately 200k bounding boxes and 880 action categories. Evaluation of existing action detection methods on FHA-Kitchens reveals varying generalization capabilities across different granularities. To handle multi-granularity in hand actions, we propose MG-HAD, an End-to-End Multi-Granularity Hand Action Detection method. It incorporates two new designs: Multi-dimensional Action Queries and Coarse-Fine Contrastive Denoising. Extensive experiments demonstrate MG-HAD's effectiveness for multi-granularity hand action detection, highlighting the significance of FHA-Kitchens for future research and real-world applications. The dataset and source code are available at MG-HAD.

Discrete diffusion models have emerged as powerful tools for high-quality data generation. Despite their success in discrete spaces, such as text generation tasks, the acceleration of discrete diffusion models remains under-explored. In this paper, we propose discrete non-Markov diffusion models (DNDM), which naturally induce the predetermined transition time set. This enables a training-free sampling algorithm that significantly reduces the number of function evaluations (i.e., calls to the neural network), making the sampling process much faster. Furthermore, we study the transition from finite to infinite step sampling, offering new insights into bridging the gap between discrete and continuous-time processes for discrete diffusion models. Extensive experiments on natural language generation and machine translation tasks demonstrate the superior performance of our method in terms of both generation speed and sample quality compared to existing methods for discrete diffusion models. Codes are available at \url{https://github.com/uclaml/DNDM}.

The research on attack transferability is of great importance as it can guide how to conduct an adversarial attack without knowing any information about target models. However, it remains challenging for adversarial examples to maintain a good attack transferability performance, especially for the black-box attack implemented in the physical world. To enhance black-box transferability of physical attacks on object detectors, we present a novel adversarial learning method to produce adversarial patches by redistributing separable attention maps. Concretely, we first develop smoothed multilayer attention maps by introducing serial composite transformations, which could suppress model-specific noise on the one hand, and cover objects to be concealed at various resolutions on the other hand. Besides, our method resorts to a scalable mask to separate object attention from the background and adjust their distribution with a novel loss function. Extensive experiments show that our approach outperforms state-of-the-art methods in both the digital space and the physical world. Our code is available at https://github.com/zhangyu13a/transPhyAtt .

A new method about renewable energy cooperation among small base stations (SBSs) is proposed, which is for maximizing the energy efficiency in ultradense network (UDN). In UDN each SBS is equipped with energy harvesting (EH) unit, and the energy arrival times are modeled as a Poisson counting process. Firstly, SBSs of large traffic demands are selected as the clustering centers, and then all SBSs are clustered using dynamic k-means algorithm. Secondly, SBSs coordinate their renewable energy within each formed cluster. The process of energy cooperation among SBSs is considered as Markov decision process. Q -learning algorithm is utilized to optimize energy cooperation. In the algorithm there are four different actions and their corresponding reward functions. Q -learning explores the action as much as possible and predicts better action by calculating reward. In addition, ε greedy policy is used to ensure the algorithm convergence. Finally, simulation results show that the new method reduces data dimension and improves calculation speed, which furthermore improves the utilization of renewable energy and promotes the performance of UDN. Through online optimization, the proposed method can significantly improve the energy utilization rate and data transmission rate.