Author name cluster

Mingda Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

AAAI Conference 2026 Conference Paper

Seeing Is Believing: Grounding Long-Video Understanding in Spatio-Temporal Visual Evidence

Zhaoyang Wei
Guoliang Wang
Guohua Gao
Yanchao Hao
Mingda Li
Wenchao Ding
Xi Chen
Shizhu He

Although Vision Language Models (VLMs) have excelled at image and video understanding, applying them to hour-long videos is held back by two interrelated challenges: exorbitant computational expense and a qualitative breakdown in long-term temporal reasoning. Thus, models tend to generate answers based on speculation instead of solid visual facts, causing both factually incorrect and plausible hallucinations. This problem is compounded by current benchmarks that, by only emphasizing final answers, lack an effective mechanism to check whether reasoning is substantiated by specific visual evidence. This makes it hard to differentiate between true understanding and pretend comprehension, inhibiting targeted model refinement. To address these interrelated challenges of model fragility and evaluation weakness, we adopt a twofold strategy. First, we present EV²-Bench, a large-scale benchmark that breaks new ground by an evaluation paradigm built upon spatio-temporal visual evidence, forcing models to justify answers with checkable hints. Second, we put forward DynamicSelect, an adaptive token compression system that efficiently condenses salient information by a dynamic semantic selector and a hierarchical compression strategy. Comprehensive experiments demonstrate that DynamicSelect significantly outperforms the baselines on EV²-Bench as well as other public benchmarks. Our study offers not only a more effective approach to long-video understanding but also a more stringent evaluation paradigm, indicating the way toward more robust models.

PDF Details DOI

JBHI Journal 2025 Journal Article

Contrastive Learning With Transformer to Predict the Chronicity of Children With Immune Thrombocytopenia

Yuntian Wang
Yongqiang Tang
Jingyao Ma
Zhenping Chen
Chang Cui
Mingda Li
Runhui Wu
Wensheng Zhang

Immune thrombocytopenia (ITP) is a typically self-limiting and immune-mediated bleeding disorder in children. Approximately 20% of children with ITP experience chronicity, leading to reduced quality of life and increased treatment burden. The accurate prediction of chronicity would enable clinicians to make personalized treatment plans at an early stage. However, due to the self-limiting nature of ITP and the scarcity of available children patients, the data presents two prominent issues: small data and imbalanced class, which are unfavorable for effectively training a deep learning model. To handle these issues concurrently, we proposed a novel method that integrates contrastive learning with the Transformer. First, we adopt the FT-Transformer as our backbone, which allows our model to flexibly process heterogeneous tabular data. Second, we amplify and balance the original data via random masking and oversampling, respectively. Lastly, we build contrastive pairs according to the latent representations generated by the FT-Transformer encoder, such that the amplified and oversampled synthetic data can be utilized thoroughly. The experimental results on real-world ITP children data show that our proposal outperforms the state-of-the-art methods, and demonstrate the significant advantages of dealing with insufficient and imbalanced problems.

Details DOI

AAAI Conference 2025 Conference Paper

TC-LLaVA: Rethinking the Transfer of LLava from Image to Video Understanding with Temporal Considerations

Mingze Gao
Jingyu Liu
Mingda Li
Jiangtao Xie
Qingbin Liu
Kevin Zhao
Xi Chen
Hui Xiong

Multimodal Large Language Models (MLLMs) have significantly improved performance across various image-language applications. Recently, there has been a growing interest in adapting image pre-trained MLLMs for video-related tasks. However, most efforts concentrate on enhancing the vision encoder and projector components, while the core part, Large Language Models (LLMs), remains comparatively under-explored. In this paper, we propose two strategies to enhance the model's capability in video understanding tasks by improving inter-layer attention computation in LLMs. Specifically, the first approach focuses on the enhancement of Rotary Position Embedding (RoPE) with Temporal-Aware Dual RoPE, which introduces temporal position information to strengthen the MLLM's temporal modeling capabilities while preserving the relative position relationships of both visual and text tokens. The second approach involves enhancing the Attention Mask with the Frame-wise Block Causal Attention Mask, a simple yet effective method that broadens visual token interactions within and across video frames while maintaining the causal inference mechanism. Based on these proposed methods, we adapt LLaVA for video understanding tasks, naming it Temporal-Considered LLaVA (TC-LLaVA). Our TC-LLaVA achieves new state-of-the-art performance across various video understanding benchmarks with only supervised fine-tuning (SFT) on video-related datasets.

PDF Details DOI

ICLR Conference 2025 Conference Paper

TRACE: Temporal Grounding Video LLM via Causal Event Modeling

Yongxin Guo 0001
Jingyu Liu
Mingda Li
Qingbin Liu
Xi Chen 0003
Xiaoying Tang 0002

Video Temporal Grounding (VTG) is a crucial capability for video understanding models and plays a vital role in downstream tasks such as video browsing and editing. To effectively handle various tasks simultaneously and enable zero-shot prediction, there is a growing trend in employing video LLMs for VTG tasks. However, current video LLM-based methods rely exclusively on natural language generation, lacking the ability to model the clear structure inherent in videos, which restricts their effectiveness in tackling VTG tasks. To address this issue, this paper first formally introduces causal event modeling framework, which represents video LLM outputs as sequences of events, and predict the current event using previous events, video inputs, and textural instructions. Each event consists of three components: timestamps, salient scores, and textual captions. We then propose a novel task-interleaved video LLM called TRACE to effectively implement the causal event modeling framework in practice. The TRACE process visual frames, timestamps, salient scores, and text as distinct tasks, employing various encoders and decoding heads for each. Task tokens are arranged in an interleaved sequence according to the causal event modeling framework's formulation. Extensive experiments on various VTG tasks and datasets demonstrate the superior performance of TRACE compared to state-of-the-art video LLMs. Our model and code are avaliable at \url{https://github.com/gyxxyg/TRACE}.

Details

AAAI Conference 2025 Conference Paper

VTG-LLM: Integrating Timestamp Knowledge into Video LLMs for Enhanced Video Temporal Grounding

Yongxin Guo
Jingyu Liu
Mingda Li
Dingxin Cheng
Xiaoying Tang
Dianbo Sui
Qingbin Liu
Xi Chen

Video Temporal Grounding (VTG) strives to accurately pinpoint event timestamps in a specific video using linguistic queries, significantly impacting downstream tasks like video browsing and editing. Unlike traditional task-specific models, Video Large Language Models (video LLMs) can handle multiple tasks concurrently in a zero-shot manner. Consequently, exploring the application of video LLMs for VTG tasks has become a burgeoning research area. However, despite considerable advancements in video content understanding, video LLMs often struggle to accurately pinpoint timestamps within videos, limiting their effectiveness in VTG tasks. To address this, we introduce VTG-LLM, a model designed to enhance video LLMs' timestamp localization abilities. Our approach includes: (1) effectively integrating timestamp knowledge into visual tokens; (2) incorporating absolute-time tokens to manage timestamp knowledge without concept shifts; and (3) introducing a lightweight, high-performance, slot-based token compression technique designed to accommodate the demands of a large number of frames to be sampled for VTG tasks. Additionally, we present VTG-IT-120K, a collection of publicly available VTG datasets that we have re-annotated to improve upon low-quality annotations. Our comprehensive experiments demonstrate the superior performance of VTG-LLM in comparison to other video LLM methods across a variety of VTG tasks.

PDF Details DOI

AAAI Conference 2024 Conference Paper

SentinelLMs: Encrypted Input Adaptation and Fine-Tuning of Language Models for Private and Secure Inference

Abhijit Mishra
Mingda Li
Soham Deo

This paper addresses the privacy and security concerns associated with deep neural language models, which serve as crucial components in various modern AI-based applications. These models are often used after being pre-trained and fine-tuned for specific tasks, with deployment on servers accessed through the internet. However, this introduces two fundamental risks: (a) the transmission of user inputs to the server via the network gives rise to interception vulnerabilities, and (b) privacy concerns emerge as organizations that deploy such models store user data with restricted context. To address this, we propose a novel method to adapt and fine-tune transformer-based language models on passkey-encrypted user-specific text. The original pre-trained language model first undergoes a quick adaptation (without any further pre-training) with a series of irreversible transformations applied to the tokenizer and token embeddings. This enables the model to perform inference on encrypted inputs while preventing reverse engineering of text from model parameters and intermediate outputs. After adaptation, models are fine-tuned on encrypted versions of existing training datasets. Experimental evaluation employing adapted versions of renowned models (e.g., BERT, RoBERTa) across established benchmark English and multilingual datasets for text classification and sequence labeling shows that encrypted models achieve performance parity with their original counterparts. This serves to safeguard performance, privacy, and security cohesively.

PDF Details DOI