Explanipedia

HKRAG: Holistic Knowledge Retrieval-Augmented Generation Over Visually-Rich Documents Open

Tong Anyang, Niu Xiang, Liu Zhiping, Tian Chang, Wei Yan-yan , et al. · 2025

Existing multimodal Retrieval-Augmented Generation (RAG) methods for visually rich documents (VRD) are often biased towards retrieving salient knowledge(e.g., prominent text and visual elements), while largely neglecting the critical fine-…

Towards Benign Memory Forgetting for Selective Multimodal Large Language Model Unlearning Open

Zeng Zhen, Gu, Leijiang, Duan Zhangling, Li Feng, Shi, Zenglin , et al. · 2025

Multimodal Large Language Models (MLLMs) achieve remarkable capabilities but can inadvertently memorize privacy-sensitive information. Although existing unlearning methods can remove such knowledge, they fail to achieve benign forgetting b…

Towards Benign Memory Forgetting for Selective Multimodal Large Language Model Unlearning Open

Zeng Zhen, Gu, Leijiang, Duan Zhangling, Li Feng, Shi, Zenglin , et al. · 2025

Multimodal Large Language Models (MLLMs) achieve remarkable capabilities but can inadvertently memorize privacy-sensitive information. Although existing unlearning methods can remove such knowledge, they fail to achieve benign forgetting b…

Harmonious Parameter Adaptation in Continual Visual Instruction Tuning for Safety-Aligned MLLMs Open

Wang ZiQi, Che Chang, Wang Qi, Ma Hui, Shi, Zenglin , et al. · 2025

While continual visual instruction tuning (CVIT) has shown promise in adapting multimodal large language models (MLLMs), existing studies predominantly focus on models without safety alignment. This critical oversight ignores the fact that…

HKRAG: Holistic Knowledge Retrieval-Augmented Generation Over Visually-Rich Documents Open

Tong Anyang, Niu Xiang, Liu Zhiping, Tian Chang, Wei Yan-yan , et al. · 2025

Existing multimodal Retrieval-Augmented Generation (RAG) methods for visually rich documents (VRD) are often biased towards retrieving salient knowledge(e.g., prominent text and visual elements), while largely neglecting the critical fine-…

Harmonious Parameter Adaptation in Continual Visual Instruction Tuning for Safety-Aligned MLLMs Open

Wang ZiQi, Che Chang, Wang Qi, Ma Hui, Shi, Zenglin , et al. · 2025

While continual visual instruction tuning (CVIT) has shown promise in adapting multimodal large language models (MLLMs), existing studies predominantly focus on models without safety alignment. This critical oversight ignores the fact that…

SAM 3: Segment Anything with Concepts Open

Carion, Nicolas, Gustafson, Laura, Hu, Yuan-Ting, Debnath, Shoubhik, Hu, Ronghang , et al. · 2025

We present Segment Anything Model (SAM) 3, a unified model that detects, segments, and tracks objects in images and videos based on concept prompts, which we define as either short noun phrases (e.g., "yellow school bus"), image exemplars,…

Development and Testing of a Modular Large-Area Cosmic Ray Telescope Using Scintillator-Fiber Hybrid Design for Millimeter-Level Muon Tracking Open

Niu Yan, Wang, Anqing, Ren, Xiangxiang, Liu Dong, Wang Meng · 2025

Cosmic-ray muons, owing to their high penetration power and abundance, have been widely employed as a natural probe in experimental particle physics. We developed a meter-scale cosmic-ray muon telescope, consisting of two parallel super-la…

SAM 3: Segment Anything with Concepts Open

Carion, Nicolas, Gustafson, Laura, Hu, Yuan-Ting, Debnath, Shoubhik, Hu, Ronghang , et al. · 2025

We present Segment Anything Model (SAM) 3, a unified model that detects, segments, and tracks objects in images and videos based on concept prompts, which we define as either short noun phrases (e.g., "yellow school bus"), image exemplars,…

Development and Testing of a Modular Large-Area Cosmic Ray Telescope Using Scintillator-Fiber Hybrid Design for Millimeter-Level Muon Tracking Open

Niu Yan, Wang, Anqing, Ren, Xiangxiang, Liu Dong, Wang Meng · 2025

Cosmic-ray muons, owing to their high penetration power and abundance, have been widely employed as a natural probe in experimental particle physics. We developed a meter-scale cosmic-ray muon telescope, consisting of two parallel super-la…

Identifying the structure of La3Ni2O7 in the pressurized superconducting state Open

Zhang Hengyuan, Zhang Jie-long, Huo, Mengwu, Chen Junfeng, HU Deyuan , et al. · 2025

The precise crystal structure of La3Ni2O7 in its high-pressure superconducting state has been a subject of intense debate, with proposed models including both orthorhombic and tetragonal symmetries. Using high-pressure Raman spectroscopy c…

An allantoin-inducible glyoxylate utilization pathway in Pseudomonas aeruginosa Open

Welch Martin, Parkhill, susannah, little, olivia, Askenasy Isabel, Labrini, edoardo , et al. · 2025

Fluorescent pseudomonads catabolize purines via uric acid and allantoin; a pathway whose end-product is glyoxylate. In this work, we show that in Pseudomonas aeruginosa strain PAO1, the open reading frames PA1498-PA1502 encode a pathway th…

Identifying the structure of La3Ni2O7 in the pressurized superconducting state Open

Zhang Hengyuan, Zhang Jie-long, Huo, Mengwu, Chen Junfeng, HU Deyuan , et al. · 2025

The precise crystal structure of La3Ni2O7 in its high-pressure superconducting state has been a subject of intense debate, with proposed models including both orthorhombic and tetragonal symmetries. Using high-pressure Raman spectroscopy c…

Probing scalar-neutrino and scalar-dark-matter interactions with PandaX-4T Open

PandaX Collaboration, Li Tao, Bo, Zihao, Chen Wei, Chen Xun , et al. · 2025

Scalar-mediated interactions may exist among neutrinos, dark matter particles, or between the two. Double $β$-decay experiments provide a powerful tool to probe such exotic interactions. Using $^{136}$Xe double $β$-decay data from PandaX-4…

Probing scalar-neutrino and scalar-dark-matter interactions with PandaX-4T Open

PandaX Collaboration, Li Tao, Bo, Zihao, Chen Wei, Chen Xun , et al. · 2025

Scalar-mediated interactions may exist among neutrinos, dark matter particles, or between the two. Double $β$-decay experiments provide a powerful tool to probe such exotic interactions. Using $^{136}$Xe double $β$-decay data from PandaX-4…

Mitigating Recommendation Biases via Group-Alignment and Global-Uniformity in Representation Learning Open

Cai Miaomiao, Hou Min, Chen Lei, Wu Le, Bai, Haoyue , et al. · 2025

Collaborative Filtering~(CF) plays a crucial role in modern recommender systems, leveraging historical user-item interactions to provide personalized suggestions. However, CF-based methods often encounter biases due to imbalances in traini…

Probing scalar-neutrino and scalar-dark-matter interactions with PandaX-4T Open

PandaX Collaboration, Li Tao, Bo, Zihao, Chen Wei, Chen Xun , et al. · 2025

Scalar-mediated interactions may exist among neutrinos, dark matter particles, or between the two. Double $β$-decay experiments provide a powerful tool to probe such exotic interactions. Using $^{136}$Xe double $β$-decay data from PandaX-4…

Mitigating Recommendation Biases via Group-Alignment and Global-Uniformity in Representation Learning Open

Cai Miaomiao, Hou Min, Chen Lei, Wu Le, Bai, Haoyue , et al. · 2025

Collaborative Filtering~(CF) plays a crucial role in modern recommender systems, leveraging historical user-item interactions to provide personalized suggestions. However, CF-based methods often encounter biases due to imbalances in traini…

Mitigating Recommendation Biases via Group-Alignment and Global-Uniformity in Representation Learning Open

Cai MiaoMiao, Hou Min, Chen Lei, Wu Le, Bai Haoyue , et al. · 2025

Collaborative Filtering~(CF) plays a crucial role in modern recommender systems, leveraging historical user-item interactions to provide personalized suggestions. However, CF-based methods often encounter biases due to imbalances in traini…

Generative Reconstruction of Spatiotemporal Wall-Pressure in Turbulent Boundary Layers via Patchwise Latent Diffusion Open

Fan, Xiantao, Parikh, Meet Hemant, Liu Yi, Liu-Xin Yang, Guo Jun-yi , et al. · 2025

Wall-pressure fluctuations in turbulent boundary layers drive flow-induced noise, structural vibration, and hydroacoustic disturbances, especially in underwater and aerospace systems. Accurate prediction of their wavenumber-frequency spect…

Generative Reconstruction of Spatiotemporal Wall-Pressure in Turbulent Boundary Layers via Patchwise Latent Diffusion Open

Fan, Xiantao, Parikh, Meet Hemant, Liu Yi, Liu-Xin Yang, Guo Jun-yi , et al. · 2025

Wall-pressure fluctuations in turbulent boundary layers drive flow-induced noise, structural vibration, and hydroacoustic disturbances, especially in underwater and aerospace systems. Accurate prediction of their wavenumber-frequency spect…

Disentangling Emotional Bases and Transient Fluctuations: A Low-Rank Sparse Decomposition Approach for Video Affective Analysis Open

Cui Feng-qi, Huang Jinyang, Jia, Ziyu, Li, Xinyu, Yan Xin , et al. · 2025

Video-based Affective Computing (VAC), vital for emotion analysis and human-computer interaction, suffers from model instability and representational degradation due to complex emotional dynamics. Since the meaning of different emotional f…

Disentangling Emotional Bases and Transient Fluctuations: A Low-Rank Sparse Decomposition Approach for Video Affective Analysis Open

Cui Feng-qi, Huang Jinyang, Jia, Ziyu, Li, Xinyu, Yan Xin , et al. · 2025

Video-based Affective Computing (VAC), vital for emotion analysis and human-computer interaction, suffers from model instability and representational degradation due to complex emotional dynamics. Since the meaning of different emotional f…

Disentangling Emotional Bases and Transient Fluctuations: A Low-Rank Sparse Decomposition Approach for Video Affective Analysis Open

Cui Feng-qi, Huang JinYang, Jia, Ziyu, Li Xinyu, Yan Xin , et al. · 2025

Video-based Affective Computing (VAC), vital for emotion analysis and human-computer interaction, suffers from model instability and representational degradation due to complex emotional dynamics. Since the meaning of different emotional f…

Supplementary materials for Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study Open

Tian Lu, Wang Che, Liu Zhihao, Du Xiaoyu, Wang Meng , et al. · 2025

AKI/PWS SR-μCT data and train/predict code Open

ZHANG Ting-Feng, Wang Ying, Chen Wei, Wang Meng, Liu, Jin-Ke , et al. · 2025

Imaging of acute kidney injury (AKI) model kidneys in mice and port-wine stain (PWS) cockscomb in chickens using SR-CTSynchrotron Radiation Micro-Computed Tomography data were acquired in absorption mode at the 4W1A beamline of the Beijing…

Supplementary materials for Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study Open

Tian Lu, Wang Che, Liu Zhi-hao, Du, Xiaoyu, Wang Meng , et al. · 2025

Supplementary materials for Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study Open

Tian Lu, Wang Che, Liu Zhihao, Du Xiaoyu, Wang Meng , et al. · 2025

Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study

Supplementary materials for Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study Open

Tian Lu, Wang Che, Liu Zhi-hao, Du, Xiaoyu, Wang Meng , et al. · 2025

Global, regional, and national burden trends of Ischemic heart disease attributable to high body mass index from 1990 to 2021, and projections to 2050: a population-based study

Towards Non-Stationary Time Series Forecasting with Temporal Stabilization and Frequency Differencing Open

Lu Junkai, Chen Peng, Guo, Chenjuan, Shu Yang, Wang Meng , et al. · 2025

Time series forecasting is critical for decision-making across dynamic domains such as energy, finance, transportation, and cloud computing. However, real-world time series often exhibit non-stationarity, including temporal distribution sh…

Wang Meng YOU? Author Swipe