Huan Xiong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Huan Xiong has a total of 97 co-authors. Here are the top co-authors by number of shared publications:
- Guo-Niu Han
I.R.M.A., UMR 7501, Université de Strasbourg et CNRS, Strasbourg, France
11 shared publications - Tian-Zhu Xiang
Inception Institute of Artificial Intelligence, Abu Dhabi, United Arab Emirates
5 shared publications - Bin Gu
Department of Machine Learning Mohamed Zayed University of Artificial Intelligence Abu Dhabi UAE
4 shared publications - Ge-Peng Ji
School of Computing, Australian National University, Canberra, Australia
4 shared publications Guo-Niu Han
Université de Strasbourg
3 shared publicationsJie Qin
Nanjing University of Aeronautics and Astronautics
3 shared publications
Latest Publications
- Top Cited Papers
- 13 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (5) , 5893-5908
VideoExpert: Augmented LLM for Temporal-Sensitive Video Understanding
The core challenge in video understanding lies in perceiving dynamic content changes over time. However, multi-modal large language models (MLLMs) struggle with temporal-sensitive video tasks, such as video temporal grounding, which requires generating timestamps to mark the occurrence of specific events. Existing strategies require MLLMs to generate absolute or relative timestamps directly. We have observed that those MLLMs tend to rely more on language patterns than visual cues when generating timestamps, affecting their performance. To address this problem, we propose VideoExpert, a general-purpose MLLM suitable for several temporal-sensitive video tasks. Inspired by the expert concept, VideoExpert integrates two parallel modules: the Temporal Expert and the Spatial Expert. The Temporal Expert is responsible for modeling time sequences and performing temporal grounding. It processes high-frame-rate yet compressed tokens to capture dynamic variations in videos and includes a lightweight prediction head for precise event localization. The Spatial Expert focuses on content detail analysis and instruction following. It handles specially designed spatial tokens and language input, aiming to generate content-related responses. These two experts collaborate seamlessly via a special token , ensuring coordinated temporal grounding and content generation. Notably, the Temporal and Spatial Experts maintain independent parameter sets. This parameter decoupling design enables specialized learning within each part without mutual interference. By offloading temporal grounding from content generation, VideoExpert prevents text pattern biases in timestamp predictions. Moreover, we introduce a Spatial Compress module to obtain spatial tokens. This module filters and compresses patch tokens while preserving key information, delivering compact yet detail-rich input for the Spatial Expert. Extensive experiments conducted on four widely-used benchmarks (i.e. Charades-STA, QVHighlight, YouCookII and NextGQA) across four tasks (temporal grounding, highlight detection, dense video captioning and grounding question answering) demonstrate the effectiveness and versatility of the VideoExpert.
- 13 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (5) , 5893-5908
Top Cited PapersVideoExpert: Augmented LLM for Temporal-Sensitive Video Understanding
The core challenge in video understanding lies in perceiving dynamic content changes over time. However, multi-modal large language models (MLLMs) struggle with temporal-sensitive video tasks, such as video temporal grounding, which requires generating timestamps to mark the occurrence of specific events. Existing strategies require MLLMs to generate absolute or relative timestamps directly. We have observed that those MLLMs tend to rely more on language patterns than visual cues when generating timestamps, affecting their performance. To address this problem, we propose VideoExpert, a general-purpose MLLM suitable for several temporal-sensitive video tasks. Inspired by the expert concept, VideoExpert integrates two parallel modules: the Temporal Expert and the Spatial Expert. The Temporal Expert is responsible for modeling time sequences and performing temporal grounding. It processes high-frame-rate yet compressed tokens to capture dynamic variations in videos and includes a lightweight prediction head for precise event localization. The Spatial Expert focuses on content detail analysis and instruction following. It handles specially designed spatial tokens and language input, aiming to generate content-related responses. These two experts collaborate seamlessly via a special token , ensuring coordinated temporal grounding and content generation. Notably, the Temporal and Spatial Experts maintain independent parameter sets. This parameter decoupling design enables specialized learning within each part without mutual interference. By offloading temporal grounding from content generation, VideoExpert prevents text pattern biases in timestamp predictions. Moreover, we introduce a Spatial Compress module to obtain spatial tokens. This module filters and compresses patch tokens while preserving key information, delivering compact yet detail-rich input for the Spatial Expert. Extensive experiments conducted on four widely-used benchmarks (i.e. Charades-STA, QVHighlight, YouCookII and NextGQA) across four tasks (temporal grounding, highlight detection, dense video captioning and grounding question answering) demonstrate the effectiveness and versatility of the VideoExpert.
- Open Access
- 1 January 2026
- Published by National Library of Serbia in Filomat
- Vol. 40 (2) , 703-719
Generalized m-Schröder paths and the Chung-Feller property
In this paper, we introduce a generalization of m-Schröder paths. For a fixed positive integer m, the generalized m-Schröder paths are lattice paths that start at (0, 0), use the steps U = (1, 1), H = (1, 0), \mathrm{V_1} = (0,-1), and \mathrm{V_2} = (0,-2) which are weighted respectively by 1, h, a and b, remain weakly above theline y =m-1/m x, and end on this line. We use generating functions and Riordan arrays to discuss the enumeration of the partial generalized m-Schröder paths and the free generalized m-Schröder paths, and obtain a Chung-Feller property. In particular, when h = a = b = 1, we find that the number of generalized m-Schröder paths of order n equals the number of hybrid (m + 1)-ary trees with n internal nodes.
- 1 January 2026
- Published by National Library of Serbia in Filomat
- Vol. 40 (2) , 703-719
Open AccessGeneralized m-Schröder paths and the Chung-Feller property
In this paper, we introduce a generalization of m-Schröder paths. For a fixed positive integer m, the generalized m-Schröder paths are lattice paths that start at (0, 0), use the steps U = (1, 1), H = (1, 0), \mathrm{V_1} = (0,-1), and \mathrm{V_2} = (0,-2) which are weighted respectively by 1, h, a and b, remain weakly above theline y =m-1/m x, and end on this line. We use generating functions and Riordan arrays to discuss the enumeration of the partial generalized m-Schröder paths and the free generalized m-Schröder paths, and obtain a Chung-Feller property. In particular, when h = a = b = 1, we find that the number of generalized m-Schröder paths of order n equals the number of hybrid (m + 1)-ary trees with n internal nodes.
- Open Access
Inverse descent statistic for André and simsun permutations
Simsun permutations, André I permutations and André II permutations are three combinatorial models for Euler numbers. It's known that the descent statistic is equidistributed over the set of André I permutations and the set of simsun permutations. In this paper, we prove that the trivariate statistic (ides, des, maj), comprising the inverse descent, descent, and major index, are equidistributed over these three sets. This result is equivalent to showing that the inverse descent is equidistributed over these three sets that share the same tree shape. The proof of the equidistribution of the inverse descent over the set of André I permutations and the set of André II permutations with the same tree shape reduces to establishing new refinements of Stanley's shuffle theorem.
Open AccessInverse descent statistic for André and simsun permutations
Simsun permutations, André I permutations and André II permutations are three combinatorial models for Euler numbers. It's known that the descent statistic is equidistributed over the set of André I permutations and the set of simsun permutations. In this paper, we prove that the trivariate statistic (ides, des, maj), comprising the inverse descent, descent, and major index, are equidistributed over these three sets. This result is equivalent to showing that the inverse descent is equidistributed over these three sets that share the same tree shape. The proof of the equidistribution of the inverse descent over the set of André I permutations and the set of André II permutations with the same tree shape reduces to establishing new refinements of Stanley's shuffle theorem.
- 23 September 2025
- Published by Springer Nature in Graphs and Combinatorics
- Vol. 41 (5) , 1-20
Generalized Narayana Polynomials and Dyck Paths with Coloured Ascents
- 23 September 2025
- Published by Springer Nature in Graphs and Combinatorics
- Vol. 41 (5) , 1-20
Generalized Narayana Polynomials and Dyck Paths with Coloured Ascents
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (3) , 1223-1237
On Expressivity of Height in Neural Networks
In this work, beyond width and depth in a traditional neural network, we augment a neural network with a new dimension called height by inserting links to neurons in the same layer, which gives rise to the notion of height. These links, referred to as intra-layer links, form hierarchical structures that enhance the network’s approximation ability (expressivity) without increasing its parameter count. We show that this 3D architecture (width × depth × height) is more powerful than conventional 2D networks (width × depth) theoretically and empirically. First, given the same number of neurons, a 3D ReLU network (W × K × H) can generate exponentially more pieces in a piecewise linear function than a 2D one (HW × K), i.e., $\mathcal{O}((2^{H} - 1)W)^{K})$ vs $\mathcal{O}((HW)^{K})$. Next, from the perspective of approximation, a 3D network (W × K × W) can achieve higher accuracy in approximating polynomials and reduce errors exponentially compared to a traditional 2D network (W2 × K), i.e., $\mathcal{O}(2^{-2W\,K})$ vs $\mathcal{O}(W^{-K})$. Lastly, numerical experiments on 5 synthetic datasets, 15 tabular datasets, and 3 image benchmarks verify that 3D networks can deliver competitive regression and classification performance. Our code has been released for readers’ free download and evaluation.
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (3) , 1223-1237
On Expressivity of Height in Neural Networks
In this work, beyond width and depth in a traditional neural network, we augment a neural network with a new dimension called height by inserting links to neurons in the same layer, which gives rise to the notion of height. These links, referred to as intra-layer links, form hierarchical structures that enhance the network’s approximation ability (expressivity) without increasing its parameter count. We show that this 3D architecture (width × depth × height) is more powerful than conventional 2D networks (width × depth) theoretically and empirically. First, given the same number of neurons, a 3D ReLU network (W × K × H) can generate exponentially more pieces in a piecewise linear function than a 2D one (HW × K), i.e., $\mathcal{O}((2^{H} - 1)W)^{K})$ vs $\mathcal{O}((HW)^{K})$. Next, from the perspective of approximation, a 3D network (W × K × W) can achieve higher accuracy in approximating polynomials and reduce errors exponentially compared to a traditional 2D network (W2 × K), i.e., $\mathcal{O}(2^{-2W\,K})$ vs $\mathcal{O}(W^{-K})$. Lastly, numerical experiments on 5 synthetic datasets, 15 tabular datasets, and 3 image benchmarks verify that 3D networks can deliver competitive regression and classification performance. Our code has been released for readers’ free download and evaluation.
- Open Access
- 16 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 19 (5) , 1-19
FDphormer: Beyond Homophily with Feature-Difference Position Encoding
Graph Transformers have garnered significant attention due to their ability to address the challenges of long-distance interactions in previous GNNs. However, most current graph Transformers face difficulties when dealing with heterophilic graphs. To investigate this issue, we first analyzed the distribution of attention weights for homophilic and heterophilic graphs. We discovered that heterophily interferes with the allocation of attention weights, leading to errors in node classification. Further investigation revealed that the root cause may be the difficulty of current graph Transformers in capturing the difference between the features of each node and its neighbors. To alleviate this issue, we propose a position encoding strategy called DiSP to better capture the feature difference and introduce FDphormer, a new efficient and simple graph Transformer model based on DiSP. Additionally, we analyze the generalization error of existing graph Transformer models and provide an upper bound on the generalization error of current graph Transformers with the introduction of DiSP. Extensive experiments demonstrate that FDphormer not only outperforms state-of-the-art methods on diverse heterogeneous datasets but also exhibits competitive performance under homophily.
- 16 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 19 (5) , 1-19
Open AccessFDphormer: Beyond Homophily with Feature-Difference Position Encoding
Graph Transformers have garnered significant attention due to their ability to address the challenges of long-distance interactions in previous GNNs. However, most current graph Transformers face difficulties when dealing with heterophilic graphs. To investigate this issue, we first analyzed the distribution of attention weights for homophilic and heterophilic graphs. We discovered that heterophily interferes with the allocation of attention weights, leading to errors in node classification. Further investigation revealed that the root cause may be the difficulty of current graph Transformers in capturing the difference between the features of each node and its neighbors. To alleviate this issue, we propose a position encoding strategy called DiSP to better capture the feature difference and introduce FDphormer, a new efficient and simple graph Transformer model based on DiSP. Additionally, we analyze the generalization error of existing graph Transformer models and provide an upper bound on the generalization error of current graph Transformers with the introduction of DiSP. Extensive experiments demonstrate that FDphormer not only outperforms state-of-the-art methods on diverse heterogeneous datasets but also exhibits competitive performance under homophily.
- Open Access
VideoExpert: Augmented LLM for Temporal-Sensitive Video Understanding
The core challenge in video understanding lies in perceiving dynamic content changes over time. However, multimodal large language models struggle with temporal-sensitive video tasks, which requires generating timestamps to mark the occurrence of specific events. Existing strategies require MLLMs to generate absolute or relative timestamps directly. We have observed that those MLLMs tend to rely more on language patterns than visual cues when generating timestamps, affecting their performance. To address this problem, we propose VideoExpert, a general-purpose MLLM suitable for several temporal-sensitive video tasks. Inspired by the expert concept, VideoExpert integrates two parallel modules: the Temporal Expert and the Spatial Expert. The Temporal Expert is responsible for modeling time sequences and performing temporal grounding. It processes high-frame-rate yet compressed tokens to capture dynamic variations in videos and includes a lightweight prediction head for precise event localization. The Spatial Expert focuses on content detail analysis and instruction following. It handles specially designed spatial tokens and language input, aiming to generate content-related responses. These two experts collaborate seamlessly via a special token, ensuring coordinated temporal grounding and content generation. Notably, the Temporal and Spatial Experts maintain independent parameter sets. By offloading temporal grounding from content generation, VideoExpert prevents text pattern biases in timestamp predictions. Moreover, we introduce a Spatial Compress module to obtain spatial tokens. This module filters and compresses patch tokens while preserving key information, delivering compact yet detail-rich input for the Spatial Expert. Extensive experiments demonstrate the effectiveness and versatility of the VideoExpert.
Open AccessVideoExpert: Augmented LLM for Temporal-Sensitive Video Understanding
The core challenge in video understanding lies in perceiving dynamic content changes over time. However, multimodal large language models struggle with temporal-sensitive video tasks, which requires generating timestamps to mark the occurrence of specific events. Existing strategies require MLLMs to generate absolute or relative timestamps directly. We have observed that those MLLMs tend to rely more on language patterns than visual cues when generating timestamps, affecting their performance. To address this problem, we propose VideoExpert, a general-purpose MLLM suitable for several temporal-sensitive video tasks. Inspired by the expert concept, VideoExpert integrates two parallel modules: the Temporal Expert and the Spatial Expert. The Temporal Expert is responsible for modeling time sequences and performing temporal grounding. It processes high-frame-rate yet compressed tokens to capture dynamic variations in videos and includes a lightweight prediction head for precise event localization. The Spatial Expert focuses on content detail analysis and instruction following. It handles specially designed spatial tokens and language input, aiming to generate content-related responses. These two experts collaborate seamlessly via a special token, ensuring coordinated temporal grounding and content generation. Notably, the Temporal and Spatial Experts maintain independent parameter sets. By offloading temporal grounding from content generation, VideoExpert prevents text pattern biases in timestamp predictions. Moreover, we introduce a Spatial Compress module to obtain spatial tokens. This module filters and compresses patch tokens while preserving key information, delivering compact yet detail-rich input for the Spatial Expert. Extensive experiments demonstrate the effectiveness and versatility of the VideoExpert.
- Open Access
A Self-Conjugate Partition Analog of $(t,t+1)$-Core Partitions with Distinct Parts
Simultaneous core partitions have been widely studied in the past 20 years. In 2013, Amdeberhan gave several conjectures on the number, the average size, and the largest size of $(t,t+1)$-core partitions with distinct parts, which was proved and generalized by Straub, Xiong, Nath-Sellers, Zaleski-Zeilberger, Paramonov, and many other mathematicians. In this paper, we introduce a proper self-conjugate partition analog of $(t,t+1)$-core partitions with distinct parts, and derive the number, the average size, and the largest size for such core partitions.
Open AccessA Self-Conjugate Partition Analog of $(t,t+1)$-Core Partitions with Distinct Parts
Simultaneous core partitions have been widely studied in the past 20 years. In 2013, Amdeberhan gave several conjectures on the number, the average size, and the largest size of $(t,t+1)$-core partitions with distinct parts, which was proved and generalized by Straub, Xiong, Nath-Sellers, Zaleski-Zeilberger, Paramonov, and many other mathematicians. In this paper, we introduce a proper self-conjugate partition analog of $(t,t+1)$-core partitions with distinct parts, and derive the number, the average size, and the largest size for such core partitions.
- 20 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 1297-1310
Deep Underwater Image Quality Assessment With Explicit Degradation Awareness Embedding
Underwater Image Quality Assessment (UIQA) is currently an area of intensive research interest. Existing deep learning-based UIQA models always learn a deep neural network to directly map the input degraded underwater image into a final quality score via end-to-end training. However, a wide variety of image contents or distortion types may correspond to the same quality score, making it challenging to train such a deep model merely with a single subjective quality score as supervision. An intuitive idea to solve this problem is to exploit more detailed degradation-aware information as supplementary guidance to facilitate model learning. In this paper, we devise a novel deep UIQA model with Explicit Degradation Awareness embedding, i.e., EDANet. To train the EDANet, a two-stage training strategy is adopted. First, a tailored Degradation Information Discovery subnetwork (DIDNet) is pre-trained to infer a residual map between the input degraded underwater image and its pseudo-reference counterpart. The inferred residual map explicitly characterizes the local degradation of the input underwater image. The intermediate feature representations on the decoder side of DIDNet are then embedded into the Degradation-guided Quality Evaluation subnetwork (DQENet), which significantly enhances the feature characterization capability with higher degradation awareness for quality prediction. The superiority of our EDANet against 18 state-of-the-art methods has been well demonstrated by extensive comparisons on two benchmark datasets. The source code of our EDANet is available at https://github.com/yia-yuese/EDANet.
- 20 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 1297-1310
Deep Underwater Image Quality Assessment With Explicit Degradation Awareness Embedding
Underwater Image Quality Assessment (UIQA) is currently an area of intensive research interest. Existing deep learning-based UIQA models always learn a deep neural network to directly map the input degraded underwater image into a final quality score via end-to-end training. However, a wide variety of image contents or distortion types may correspond to the same quality score, making it challenging to train such a deep model merely with a single subjective quality score as supervision. An intuitive idea to solve this problem is to exploit more detailed degradation-aware information as supplementary guidance to facilitate model learning. In this paper, we devise a novel deep UIQA model with Explicit Degradation Awareness embedding, i.e., EDANet. To train the EDANet, a two-stage training strategy is adopted. First, a tailored Degradation Information Discovery subnetwork (DIDNet) is pre-trained to infer a residual map between the input degraded underwater image and its pseudo-reference counterpart. The inferred residual map explicitly characterizes the local degradation of the input underwater image. The intermediate feature representations on the decoder side of DIDNet are then embedded into the Degradation-guided Quality Evaluation subnetwork (DQENet), which significantly enhances the feature characterization capability with higher degradation awareness for quality prediction. The superiority of our EDANet against 18 state-of-the-art methods has been well demonstrated by extensive comparisons on two benchmark datasets. The source code of our EDANet is available at https://github.com/yia-yuese/EDANet.
- Open Access
- 28 June 2024
- Published by Springer Nature in Visual Intelligence
- Vol. 2 (1) , 1-17
Effectiveness assessment of recent large vision-language models
The advent of large vision-language models (LVLMs) represents a remarkable advance in the quest for artificial general intelligence. However, the models’ effectiveness in both specialized and general tasks warrants further investigation. This paper endeavors to evaluate the competency of popular LVLMs in specialized and general tasks, respectively, aiming to offer a comprehensive understanding of these novel models. To gauge their effectiveness in specialized tasks, we employ six challenging tasks in three different application scenarios: natural, healthcare, and industrial. These six tasks include salient/camouflaged/transparent object detection, as well as polyp detection, skin lesion detection, and industrial anomaly detection. We examine the performance of three recent open-source LVLMs, including MiniGPT-v2, LLaVA-1.5, and Shikra, on both visual recognition and localization in these tasks. Moreover, we conduct empirical investigations utilizing the aforementioned LVLMs together with GPT-4V, assessing their multi-modal understanding capabilities in general tasks including object counting, absurd question answering, affordance reasoning, attribute recognition, and spatial relation reasoning. Our investigations reveal that these LVLMs demonstrate limited proficiency not only in specialized tasks but also in general tasks. We delve deep into this inadequacy and uncover several potential factors, including limited cognition in specialized tasks, object hallucination, text-to-image interference, and decreased robustness in complex problems. We hope that this study can provide useful insights for the future development of LVLMs, helping researchers improve LVLMs for both general and specialized applications.
- 28 June 2024
- Published by Springer Nature in Visual Intelligence
- Vol. 2 (1) , 1-17
Open AccessEffectiveness assessment of recent large vision-language models
The advent of large vision-language models (LVLMs) represents a remarkable advance in the quest for artificial general intelligence. However, the models’ effectiveness in both specialized and general tasks warrants further investigation. This paper endeavors to evaluate the competency of popular LVLMs in specialized and general tasks, respectively, aiming to offer a comprehensive understanding of these novel models. To gauge their effectiveness in specialized tasks, we employ six challenging tasks in three different application scenarios: natural, healthcare, and industrial. These six tasks include salient/camouflaged/transparent object detection, as well as polyp detection, skin lesion detection, and industrial anomaly detection. We examine the performance of three recent open-source LVLMs, including MiniGPT-v2, LLaVA-1.5, and Shikra, on both visual recognition and localization in these tasks. Moreover, we conduct empirical investigations utilizing the aforementioned LVLMs together with GPT-4V, assessing their multi-modal understanding capabilities in general tasks including object counting, absurd question answering, affordance reasoning, attribute recognition, and spatial relation reasoning. Our investigations reveal that these LVLMs demonstrate limited proficiency not only in specialized tasks but also in general tasks. We delve deep into this inadequacy and uncover several potential factors, including limited cognition in specialized tasks, object hallucination, text-to-image interference, and decreased robustness in complex problems. We hope that this study can provide useful insights for the future development of LVLMs, helping researchers improve LVLMs for both general and specialized applications.