Xingjian Diao

Whole slide images, with their gigapixel-scale panoramas of tissue samples, are pivotal for precise disease diagnosis. However, their analysis is hindered by immense data size and scarce annotations. Existing MIL methods face challenges due to the fundamental imbalance where a single bag-level label must guide the learning of numerous patch-level features. This sparse supervision makes it difficult to reliably identify discriminative patches during training, leading to unstable optimization and suboptimal solutions. We propose a spatially regularized MIL framework that leverages inherent spatial relationships among patch features as label-independent regularization signals. Our approach learns a shared representation space by jointly optimizing feature-induced spatial reconstruction and label-guided classification objectives, enforcing consistency between intrinsic structural patterns and supervisory signals. Experimental results on multiple public datasets demonstrate significant improvements over state-of-the-art methods, offering a promising direction.

As large language models (LLMs) are evolving at a rapid pace, the design space for building high-performance and efficient models has expanded significantly. This position paper argues that the next level of LLMs will combine decoder-only architectures with modular enhancements in perception and specialization, supported by adaptive post-training and hardware-aware optimization strategies. We analyze 31 influential LLMs developed by 12 leading institutions, examining their design choices across three critical dimensions: model architecture, post-training, and optimization. By asking and answering 9 fundamental research questions surrounding these design choices, we identify key trends and propose forward-looking directions to guide the design of next-generation LLMs that are more capable, efficient, and adaptable to diverse real-world needs. To support the community, we maintain a GitHub repository that will be continuously updated at https://github.com/meettyj/LLMs-Design-Choices .

Music performances, characterized by dense and continuous audio as well as seamless audio-visual integration, present unique challenges for multimodal scene understanding and reasoning. Recent Music Performance Audio-Visual Question Answering (Music AVQA) datasets have been proposed to reflect these challenges, highlighting the continued need for more effective integration of audio-visual representations in complex question answering. However, existing Music AVQA methods often rely on dense and unoptimized representations, leading to inefficiencies in the isolation of key information, the reduction of redundancy, and the prioritization of critical samples. To address these challenges, we introduce Sparsify, a sparse learning framework specifically designed for Music AVQA. It integrates three sparsification strategies into an end-to-end pipeline and achieves state-of-the-art performance on the Music AVQA datasets. In addition, it reduces training time by 28.32% compared to its fully trained dense counterpart while maintaining accuracy, demonstrating clear efficiency gains. To further improve data efficiency, we propose a key-subset selection algorithm that selects and uses approximately 25% of MUSIC-AVQA v2.0 training data and retains 70-80% of full-data performance across models.

Music performances are representative scenarios for audio-visual modeling. Unlike common scenarios with sparse audio, music performances continuously involve dense audio signals throughout. While existing multimodal learning methods on the audio-video QA demonstrate impressive capabilities in general scenarios, they are incapable of dealing with fundamental problems within the music performances: they underexplore the interaction between the multimodal signals in performance and fail to consider the distinctive characteristics of instruments and music. Therefore, existing methods tend to answer questions regarding musical performances inaccurately. To bridge the above research gaps, (i) given the intricate multimodal interconnectivity inherent to music data, our primary backbone is designed to incorporate multimodal interactions within the context of music; (ii) to enable the model to learn music characteristics, we annotate and release rhythmic and music sources in the current music datasets; (iii) for time-aware audio-visual modeling, we align the model's music predictions with the temporal dimension. Our experiments show state-of-the-art effects on the Music AVQA datasets. Our code is available at https://github.com/xid32/Amuse.

Xingjian Diao, Chunhui Zhang, Keyi Kong, Weiyi Wu, Chiyu Ma, Zhongyu Ouyang, Peijun Qing, Soroush Vosoughi, Jiang Gui. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Chunhui Zhang, Zhongyu Ouyang, Xingjian Diao, Zheyuan Liu, Soroush Vosoughi. Findings of the Association for Computational Linguistics: EMNLP 2025. 2025.

Xingjian Diao, Tianzhen Yang, Chunhui Zhang, Weiyi Wu, Ming Cheng, Jiang Gui. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2025.

Due to the challenges in mathematical solving, accurate analytical solutions for rectangular thin plates are mainly limited to these plates subject to simple Lévy-type boundary conditions (BCs). Conversely, plates with more complicated BCs are more frequently encountered in practical engineering applications. This study employs the finite integral transform method to analyze the free vibration behavior of isotropic rectangular thin plates with three edges rotationally-restrained and one edge free, which is difficult for other methods. Primarily, the method adopted eliminates the necessity for predetermining trial function for plate deflection, offering simplicity and generality for plate mechanical problems. Furthermore, the approach establishes one way to allow one solving linear algebraic equations instead of handling complex boundary value problems of higher-order vibration partial differential equation (PDE), resulting precise and rapidly converging analytical solutions. Finally, new accurate free vibration studies for plates with four non-Lévy-type BCs and three non-classical BCs are presented via altering the introduced rotating fixed coefficients. The results obtained are contrasted with the numerical simulations from ABAQUS software and existing literature works, confirming the accuracy of this method and providing benchmark results for further exploration of free vibration problems of rectangular plates.