Zehao Li

EAAI Journal 2026 Journal Article

Two level time-aware network for clinical event prediction

• Qing Li
• Zehao Li

AAAI Conference 2025 Conference Paper

Attributive Reasoning for Hallucination Diagnosis of Large Language Models

• Yuyan Chen
• Zehao Li
• Shuangjie You
• Zhengyu Chen
• Jingwen Chang
• Yi Zhang
• Weinan Dai
• Qingpei Guo

In recent years, large language models (LLMs) have demonstrated outstanding capabilities in various tasks. However, LLMs also have various drawbacks, especially hallucination. Hallucination refers to the generation of content that does not align with the user input, contradicts previously generated content or world knowledge. Current research on hallucination mainly include knowledge retrieval, prompt engineering, training data improvement, reinforcement learning, etc. However, these methods do not involve different categories of hallucinations which is important on hallucination analysis, and make detailed investigation for the internal state of LLMs which indicates the direction on hallucination occurrence. Therefore, in our research, we introduce an attribution framework to trace the origins of hallucinations based on the internal signals of LLMs. To support this framework, we develop a new benchmark named RelQA-Cate, which includes eight categories of hallucinations for the answers generated by LLMs. After that, we present a novel Differential Penalty Decoding (DPD) strategy for reducing hallucinations through adjusting post-probabilities of each answer. We conduct a series of experiments and the performance on answer reliability has significant improvement, achieving 28.25% at most, which demonstrates the effectiveness of our proposed DPD and its generalization in mitigating hallucination in LLMs.

NeurIPS Conference 2025 Conference Paper

HAIF-GS: Hierarchical and Induced Flow-Guided Gaussian Splatting for Dynamic Scene

• Jianing Chen
• Zehao Li
• Yujun Cai
• Hao Jiang
• Chengxuan Qian
• Juyuan Kang
• Shuqin Gao
• Honglong Zhao

Reconstructing dynamic 3D scenes from monocular videos remains a fundamental challenge in 3D vision. While 3D Gaussian Splatting (3DGS) achieves real-time rendering in static settings, extending it to dynamic scenes is challenging due to the difficulty of learning structured and temporally consistent motion representations. This challenge often manifests as three limitations in existing methods: redundant Gaussian updates, insufficient motion supervision, and weak modeling of complex non-rigid deformations. These issues collectively hinder coherent and efficient dynamic reconstruction. To address these limitations, we propose HAIF-GS, a unified framework that enables structured and consistent dynamic modeling through sparse anchor-driven deformation. It first identifies motion-relevant regions via an Anchor Filter to suppress redundant updates in static areas. A self-supervised Induced Flow-Guided Deformation module induces anchor motion using multi-frame feature aggregation, eliminating the need for explicit flow labels. To further handle fine-grained deformations, a Hierarchical Anchor Propagation mechanism increases anchor resolution based on motion complexity and propagates multi-level transformations. Extensive experiments on synthetic and real-world benchmarks validate that HAIF-GS significantly outperforms prior dynamic 3DGS methods in rendering quality, temporal coherence, and reconstruction efficiency.

IROS Conference 2025 Conference Paper

TBAP: Tapping-Based Auditory Perception for Identifying Container Materials

• Zehao Li
• Shoujie Li
• Hao Guo
• Wenbo Ding

In this study, in order to address the robotic auditory perception problem, we propose a novel framework for object material recognition of common containers, which combines deep learning with active auditory perception to achieve breakthrough results. We developed a modular robotic system for acoustic data acquisition that employs a hybrid mechanism of vertical translation and horizontal rotation that is capable of performing full-scale tapping in three dimensions. The system is capable of creating an acoustic dataset consisting of 50 containers made of five materials, which improves the data acquisition efficiency by 93. 9% compared to manual operations. In addition, we propose an end-to-end transfer learning model, TBAP, which is trained on a crawler-generated pre-training dataset and 50 real scene samples, and achieves a recognition accuracy of 91. 0% for unseen materials. To improve reliability, we design a dynamic confidence assessment mechanism that generates confidence indices through probability distribution analysis and feature stability assessment to support robust robot decision-making. Experimental results show that the framework greatly improves data acquisition efficiency while maintaining high recognition accuracy, providing a valuable tool for advancing acoustic perception research.

NeurIPS Conference 2025 Conference Paper

Unbalanced Optimal Total Variation Transport: A Theoretical Approach to Spatial Resource Allocation Problems

• Nhan-Phu Chung
• Jinhui Han
• Bohan Li
• Zehao Li

We propose and analyze a new class of unbalanced weak optimal transport (OT) problems with total variation penalties, motivated by spatial resource allocation tasks. Unlike classical OT, our framework accommodates general unbalanced nonnegative measures and incorporates cost objectives that directly capture operational trade-offs between transport cost and supply–demand mismatch. In the general setting, we establish the existence of optimal solutions and a dual formulation. We then focus on the semi-discrete setting, where one measure is discrete and the other is absolutely continuous, a structure relevant to applications such as service area partitioning for facilities like schools or medical stations. Exploiting a tessellation-based structure, we derive the corresponding explicit optimality conditions. We further address a quantization problem that jointly optimizes the locations and weights of discrete support points, applicable to facility location tasks such as the cost-efficient deployment of battery swap stations or e-commerce warehouses, informed by demand-side data. The dual-tessellation structure also yields explicit gradient expressions, enabling efficient numerical optimization in finite dimensions.