Spatial learning is a crucial cognitive process through which individuals acquire and retain information about their environment, playing a vital role in real-world applications such as emergency training and urban planning. Recent research has increasingly focused on virtual environments, where visualizations such as path encoding can significantly aid in understanding spatial knowledge, such as escape routes in emergency situations. To explore how different virtual environments and perspectives enhance spatial learning under visualization, this study investigated two different virtual environments — desktop and Virtual Reality (VR) — and the impacts of first-person and third-person perspectives. We developed a prototype based on expert interviews and real-world needs, conducting a between-subjects study involving 48 participants. The experimental design systematically varied the virtual environments and perspectives to assess their effects on participants’ performance of spatial learning under the same visual cues. The focus was on three types of spatial knowledge: landmark knowledge, route knowledge, and survey knowledge. The results indicate that, under the same virtual environment, the third-person perspective is more effective in enhancing survey knowledge, while the first-person perspective is better suited for acquiring landmark knowledge. Furthermore, using the third-person perspective in VR environments and the first-person perspective in desktop environments yields better spatial learning outcomes compared to other conditions.

Yuyao Ge, Shenghua Liu, Baolong Bi, Yiwei Wang, Lingrui Mei, Wenjie Feng, Lizhe Chen, Xueqi Cheng. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

With the swift progression of image generation technology, the widespread emergence of facial deepfakes poses significant challenges to the field of security, thus amplifying the urgent need for effective deepfake detection. Existing techniques for face forgery detection can broadly be categorized into two primary groups: visual-based methods and multimodal approaches. The former often lacks clear explanations for forgery details, while the latter, which merges visual and linguistic modalities, is more prone to the issue of hallucinations.To address these shortcomings, we introduce a visual detail enhanced self-correction framework, designated CorrDetail, for interpretable face forgery detection. CorrDetail is meticulously designed to rectify authentic forgery details when provided with error-guided questioning, with the aim of fostering the ability to uncover forgery details rather than yielding hallucinated responses. Additionally, to bolster the reliability of its findings, a visual fine-grained detail enhancement module is incorporated, supplying CorrDetail with more precise visual forgery details. Ultimately, a fusion decision strategy is devised to further augment the model's discriminative capacity in handling extreme samples, through the integration of visual information compensation and model bias reduction. Experimental results demonstrate that CorrDetail not only achieves state-of-the-art performance compared to the latest methodologies but also excels in accurately identifying forged details, all while exhibiting robust generalization capabilities.

Developing agents with ability to simulate human emotional processes is a key aspect of AGI, but most current work focuses on optimizing the decision-making performance of Large Language models (LLMs). To align the desicions of absolutely rational LLMs with the behavioral logic of humans characterized by emotional traits, we propose FEMA (Feeling-Emotional-Moody Agent), a novel framework for agents to achieve human-like personification by integrating emotional dynamics. FEMA consists of three core components: Emotion Perceiver, which generates emotional responses from input stimuli; Emotion Driver, which guides authentic emotional reactions and behavior; and Emotion Fluctuation Strategy, which introduces natural emotional variability to mimic human emotional unpredictability. We evaluate our approach on a virtual social platform, where agents participate in discussions on various topics. Results show that FEMA maintains logical coherence while exhibiting rich emotional expression, leading to more realistic and balanced personification. The code is available as open-source at https://github.com/Xiaoshaoxinxing/FEMA.

Abstract Adversarial attack for time-series classification model is widely explored and many attack methods are proposed. But there is not a method of attack based on the data itself. In this paper, we innovatively proposed a black-box sparse attack method based on data location. Our method directly attack the sensitive points in the time-series data according to statistical features extract from the dataset. At first, we have validated the transferability of sensitive points among DNNs with different structures. Secondly, we use the statistical features extract from the dataset and the sensitive rate of each point as the training set to train the predictive model. Then, predicting the sensitive rate of test set by predictive model. Finally, perturbing according to the sensitive rate. The attack is limited by constraining the L0 norm to achieve one-point attack. We conduct experiments on several datasets to validate the effectiveness of this method.

Recent self-supervised video representation learning methods have found significant success by exploring essential properties of videos, e.g. speed, temporal order, etc. This work exploits an essential yet under-explored property of videos, the \textit{video continuity}, to obtain supervision signals for self-supervised representation learning. Specifically, we formulate three novel continuity-related pretext tasks, i.e. continuity justification, discontinuity localization, and missing section approximation, that jointly supervise a shared backbone for video representation learning. This self-supervision approach, termed as Continuity Perception Network (CPNet), solves the three tasks altogether and encourages the backbone network to learn local and long-ranged motion and context representations. It outperforms prior arts on multiple downstream tasks, such as action recognition, video retrieval, and action localization. Additionally, the video continuity can be complementary to other coarse-grained video properties for representation learning, and integrating the proposed pretext task to prior arts can yield much performance gains.