Min Xia

Jian Li, Weiheng Lu, Hao Fei et al.

Multimodal Large Language Models (MLLMs) are gaining increasing popularity in both academia and industry due to their remarkable performance in various applications such as visual question answering, visual perception, understanding, and reasoning. Over the past few years, significant efforts have been made to examine MLLMs from multiple perspectives. This paper presents a comprehensive review of 200 benchmarks and evaluations for MLLMs, focusing on (1)perception and understanding, (2)cognition and reasoning, (3)specific domains, (4)key capabilities, and (5)other modalities. Finally, we discuss the limitations of the current evaluation methods for MLLMs and explore promising future directions. Our key argument is that evaluation should be regarded as a crucial discipline to support the development of MLLMs better. For more details, please visit our GitHub repository: https://github.com/swordlidev/Evaluation-Multimodal-LLMs-Survey.

Jingbin Qian, Congwen Yi, Min Xia et al.

Retrieval-Augmented Generation (RAG) has become critical for knowledge-intensive applications, yet evaluating its performance in vertical domains remains difficult due to domain complexity, diverse context scales, and heavy reliance on expert assessments that are costly, inconsistent, and non-scalable. We introduce FAB-Bench, an end-to-end framework for adaptive benchmarking of RAG systems in semiconductor manufacturing. FAB-Bench defines six diagnostic metrics measuring factual accuracy, contextual utilization, completeness, retrieval relevance, technical depth, and reasoning consistency. The framework couples retriever diagnostics with generator-level reasoning analysis across context windows of 4K-32K tokens, quantifying how retrieval precision and generative fidelity co-evolve as contextual scope expands. From over 1,300 generated candidates, we curated a high-quality benchmark of 200 query-answer pairs spanning three synthesis strategies: needle-in-haystack, intra-document multi-topic, and cross-document multi-hop. Systematic evaluation across four LLMs and four RAG frameworks reveals three distinct context-scaling behaviors: logarithmic growth, early saturation, and cold-start dynamics, and identifies attention dilution as the primary mechanism behind performance degradation at extreme context lengths. Cross-framework validation on three additional production RAG systems confirms evaluation portability.

Weiheng Lu, Jian Li, An Yu et al.

Multimodal Large Language Models (MLLMs) are widely used for visual perception, understanding, and reasoning. However, long video processing and precise moment retrieval remain challenging due to LLMs' limited context size and coarse frame extraction. We propose the Large Language-and-Vision Assistant for Moment Retrieval (LLaVA-MR), which enables accurate moment retrieval and contextual grounding in videos using MLLMs. LLaVA-MR combines Dense Frame and Time Encoding (DFTE) for spatial-temporal feature extraction, Informative Frame Selection (IFS) for capturing brief visual and motion patterns, and Dynamic Token Compression (DTC) to manage LLM context limitations. Evaluations on benchmarks like Charades-STA and QVHighlights demonstrate that LLaVA-MR outperforms 11 state-of-the-art methods, achieving an improvement of 1.82% in R1@0.5 and 1.29% in mAP@0.5 on the QVHighlights dataset. Our implementation will be open-sourced upon acceptance.

David Li, Nels Numan, Xun Qian et al.

We are on the cusp where Artificial Intelligence (AI) and Extended Reality (XR) are converging to unlock new paradigms of interactive computing. However, a significant gap exists between the ecosystems of these two fields: while AI research and development is accelerated by mature frameworks like JAX and benchmarks like LMArena, prototyping novel AI-driven XR interactions remains a high-friction process, often requiring practitioners to manually integrate disparate, low-level systems for perception, rendering, and interaction. To bridge this gap, we present XR Blocks, a cross-platform framework designed to accelerate human-centered AI + XR innovation. XR Blocks strives to provide a modular architecture with plug-and-play components for core abstraction in AI + XR: user, world, peers; interface, context, and agents. Crucially, it is designed with the mission of "reducing frictions from idea to reality", thus accelerating rapid prototyping of AI + XR apps. Built upon accessible technologies (WebXR, three.js, TensorFlow, Gemini), our toolkit lowers the barrier to entry for XR creators. We demonstrate its utility through a set of open-source templates, samples, and advanced demos, empowering the community to quickly move from concept to interactive XR prototype. Site: https://xrblocks.github.io

Feilong Jiang, Xiaonan Hou, Min Xia

Predicting the Remaining Useful Life (RUL) is essential in Prognostic Health Management (PHM) for industrial systems. Although deep learning approaches have achieved considerable success in predicting RUL, challenges such as low prediction accuracy and interpretability pose significant challenges, hindering their practical implementation. In this work, we introduce a Spatio-temporal Attention-based Hidden Physics-informed Neural Network (STA-HPINN) for RUL prediction, which can utilize the associated physics of the system degradation. The spatio-temporal attention mechanism can extract important features from the input data. With the self-attention mechanism on both the sensor dimension and time step dimension, the proposed model can effectively extract degradation information. The hidden physics-informed neural network is utilized to capture the physics mechanisms that govern the evolution of RUL. With the constraint of physics, the model can achieve higher accuracy and reasonable predictions. The approach is validated on a benchmark dataset, demonstrating exceptional performance when compared to cutting-edge methods, especially in the case of complex conditions.

Feilong Jiang, Xiaonan Hou, Min Xia

Although physics-informed neural networks (PINNs) have shown great potential in dealing with nonlinear partial differential equations (PDEs), it is common that PINNs will suffer from the problem of insufficient precision or obtaining incorrect outcomes. Unlike most of the existing solutions trying to enhance the ability of PINN by optimizing the training process, this paper improved the neural network architecture to improve the performance of PINN. We propose a densely multiply PINN (DM-PINN) architecture, which multiplies the output of a hidden layer with the outputs of all the behind hidden layers. Without introducing more trainable parameters, this effective mechanism can significantly improve the accuracy of PINNs. The proposed architecture is evaluated on four benchmark examples (Allan-Cahn equation, Helmholtz equation, Burgers equation and 1D convection equation). Comparisons between the proposed architecture and different PINN structures demonstrate the superior performance of the DM-PINN in both accuracy and efficiency.

Feilong Jiang, Xiaonan Hou, Jianqiao Ye et al.

Physics-Informed Neural Networks (PINNs) have emerged as a powerful framework for solving partial differential equations (PDEs) by embedding physical laws directly into the loss function. However, as a fundamental optimization issue, internal covariate shift (ICS) hinders the stable and effective training of PINNs by disrupting feature distributions and limiting model expressiveness. Unlike standard deep learning tasks, conventional remedies for ICS -- such as Batch Normalization and Layer Normalization -- are not directly applicable to PINNs, as they distort the physical consistency required for reliable PDE solutions. To address this issue, we propose Mask-PINNs, a novel architecture that introduces a learnable mask function to regulate feature distributions while preserving the underlying physical constraints of PINNs. We provide a theoretical analysis showing that the mask suppresses the expansion of feature representations through a carefully designed modulation mechanism. Empirically, we validate the method on multiple PDE benchmarks -- including convection, wave propagation, and Helmholtz equations -- across diverse activation functions. Our results show consistent improvements in prediction accuracy, convergence stability, and robustness. Furthermore, we demonstrate that Mask-PINNs enable the effective use of wider networks, overcoming a key limitation in existing PINN frameworks.

Zihan Zhou, Ziyi Zeng, Wenhao Jiang et al.

As social changes accelerate, the incidence of psychosomatic disorders has significantly increased, becoming a major challenge in global health issues. This necessitates an innovative knowledge system and analytical methods to aid in diagnosis and treatment. Here, we establish the ontology model and entity types, using the BERT model and LoRA-tuned LLM for named entity recognition, constructing the knowledge graph with 9668 triples. Next, by analyzing the network distances between disease, symptom, and drug modules, it was found that closer network distances among diseases can predict greater similarities in their clinical manifestations, treatment approaches, and psychological mechanisms, and closer distances between symptoms indicate that they are more likely to co-occur. Lastly, by comparing the proximity d and proximity z score, it was shown that symptom-disease pairs in primary diagnostic relationships have a stronger association and are of higher referential value than those in diagnostic relationships. The research results revealed the potential connections between diseases, co-occurring symptoms, and similarities in treatment strategies, providing new perspectives for the diagnosis and treatment of psychosomatic disorders and valuable information for future mental health research and practice.

Feilong Jiang, Xiaonan Hou, Min Xia

As a promising framework for resolving partial differential equations (PDEs), Physics-Informed Neural Networks (PINNs) have received widespread attention from industrial and scientific fields. However, lack of expressive ability and initialization pathology issues are found to prevent the application of PINNs in complex PDEs. In this work, we propose Deeper Physics-Informed Neural Network (Deeper-PINN) to resolve these issues. The element-wise multiplication operation is adopted to transform features into high-dimensional, non-linear spaces. Benefiting from element-wise multiplication operation, Deeper-PINNs can alleviate the initialization pathologies of PINNs and enhance the expressive capability of PINNs. The proposed structure is verified on various benchmarks. The results show that Deeper-PINNs can effectively resolve the initialization pathology and exhibit strong expressive ability.

Ming Qian, Min Xia, Chunyi Sun et al.

Defocus blur always occurred in photos when people take photos by Digital Single Lens Reflex Camera(DSLR), giving salient region and aesthetic pleasure. Defocus blur Detection aims to separate the out-of-focus and depth-of-field areas in photos, which is an important work in computer vision. Current works for defocus blur detection mainly focus on the designing of networks, the optimizing of the loss function, and the application of multi-stream strategy, meanwhile, these works do not pay attention to the shortage of training data. In this work, to address the above data-shortage problem, we turn to rethink the relationship between two tasks: defocus blur detection and salient region detection. In an image with bokeh effect, it is obvious that the salient region and the depth-of-field area overlap in most cases. So we first train our network on the salient region detection tasks, then transfer the pre-trained model to the defocus blur detection tasks. Besides, we propose a novel network for defocus blur detection. Experiments show that our transfer strategy works well on many current models, and demonstrate the superiority of our network.

Ming Liu, Yukang Ding, Min Xia et al.

Arbitrary attribute editing generally can be tackled by incorporating encoder-decoder and generative adversarial networks. However, the bottleneck layer in encoder-decoder usually gives rise to blurry and low quality editing result. And adding skip connections improves image quality at the cost of weakened attribute manipulation ability. Moreover, existing methods exploit target attribute vector to guide the flexible translation to desired target domain. In this work, we suggest to address these issues from selective transfer perspective. Considering that specific editing task is certainly only related to the changed attributes instead of all target attributes, our model selectively takes the difference between target and source attribute vectors as input. Furthermore, selective transfer units are incorporated with encoder-decoder to adaptively select and modify encoder feature for enhanced attribute editing. Experiments show that our method (i.e., STGAN) simultaneously improves attribute manipulation accuracy as well as perception quality, and performs favorably against state-of-the-arts in arbitrary facial attribute editing and season translation.