Author name cluster

Li Kong

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.

5 papers

1 author row

AAAI Conference 2026 Conference Paper

SACO: Sequence-Aware Constrained Optimization Framework for Coupon Distribution in E-commerce

Li Kong
Bingzhe Wang
Zhou Chen
Suhan Hu
Yuchao Ma
Qi Qi
Suoyuan Song
Bicheng Jin

Coupon distribution is a critical marketing strategy used by online platforms to boost revenue and enhance user engagement. Regrettably, existing coupon distribution strategies fall far short of effectively leveraging the complex sequential interactions between platforms and users. This critical oversight, despite the abundance of e-commerce log data, has precipitated a performance plateau. In this paper, we focus on the scene that the platforms make sequential coupon distribution decision multiple times for various users, with each user interacting with the platform repeatedly. Based on this marketing scenario, we propose a novel marketing framework, named Sequence-Aware Constrained Optimization (SACO) framework, to directly devise coupon distribution policy for long-term revenue boosting. SACO framework enables optimized online decision-making in a variety of real-world marketing scenarios. It achieves this by seamlessly integrating three key characteristics, general scenarios, sequential modeling with more comprehensive historical data, and efficient iterative updates within a unified framework. Furthermore, empirical results on real-world industrial dataset, alongside public and synthetic datasets demonstrate the superiority of our framework.

PDF Details DOI

YNIMG Journal 2025 Journal Article

Lateralization study of the basal ganglia, thalamus and supplying arteries in healthy individuals based on structure and connectivity analysis using 7.0T MRI

Hongqin Liang
Yawei Gu
Xiaoqi Yi
Li Kong
Jian Wang
Fajin Lv

The human cerebral cortex is known for its hemispheric specialization, which underpins a variety of functions and activities. However, it is not well understood if similar lateralization exists within the deep gray matter nuclei, such as the basal ganglia (BG) and thalamus, and their associated arteries, including the lenticulostriate arteries (LSAs). To explore this, we analyzed images from 7T MRI scans of 40 healthy young individuals. We isolated the LSAs and analyzed their morphological characteristics. Additionally, we segmented the bilateral BG and thalami into 28 subregions based on the Brainnetome atlas and calculated their volumes using single-subject voxel-based morphometry (VBM) analysis. We also obtained four parameters from Diffusion Tensor Imaging (DTI) within these 28 subregions. Bilateral comparisons were conducted using paired t-tests for all measurements. Connectivity-network matrices, based on the number of connecting fibers and fractional anisotropy between subregions, were constructed. The results showed that in the right-handed dominant population, we observed that the total number and length of LSAs in the left hemisphere exceeded those in the right. Among the 28 subregions, several showed leftward volume dominance, including the ventral caudate, globus pallidus, ventromedial putamen, medial pre-frontal thalamus, occipital thalamus, and caudal temporal thalamus. In contrast, the nucleus accumbens, dorsal caudate, dorsolateral putamen, pre-motor thalamus, sensory thalamus, posterior parietal thalamus, and lateral pre-frontal thalamus showed rightward volume dominance. Except for the ventral medial putamen (vmPu), all other subnuclei displayed at least three DTI measures with left-right differences. The connectivity between subregions revealed distinct patterns in the bilateral hemispheres, with a leftward dominance. Collectively, these findings enhance our understanding of lateralization within subcortical structures.

Details DOI

YNIMG Journal 2025 Journal Article

Predicting cognitive aging through brain structural covariance networks: A decade of longitudinal insights using source-based morphometry

Xingsong Wang
Christina J. Herold
Li Kong
Raymond C.K. Chan
Johannes Schröder

Cognitive aging presents significant challenges to public health as the global population ages. While functional connectivity changes in aging have been extensively studied, the predictive value of structural covariance networks remains understudied. This longitudinal study investigated whether baseline structural covariance networks could predict cognitive decline over a 10-year period using Source-Based Morphometry (SBM). Thirty-seven participants (23 males; mean age 54.97 ± 1.14 years) underwent structural magnetic resonance imaging (T3) and cognitive assessments at baseline (T3) and follow-up (T4). SBM analysis identified twelve independent components (ICs) representing distinct structural covariance networks. After controlling for demographics and APOE genotype, IC1 strongly predicted working memory (β = -3.12, p< 0.001), while IC2 predicted global cognitive function (β = 0.37, p= 0.047). Brain-cognition relationships were significantly moderated by baseline cognitive performance, with key interactions observed for working memory and IC1 (β = 0.50, p< 0.001), executive function and IC7 (β = -0.25, p< 0.001), and processing speed and IC8 (β = 0.28, p= 0.003). Sex-specific effects emerged for IC8 in relation to verbal memory (β = 1.99, p= 0.007) and IC10 in relation to processing speed (β = 2.17, p= 0.022). APOE genotype demonstrated pronounced moderation effects between IC8 and processing speed (β = -7.68, p< 0.001) and for IC2 and global cognitive function (β = 0.37, p= 0.018). These findings demonstrate that structural covariance networks can serve as predictive markers for cognitive aging trajectories, potentially informing early intervention strategies for preserving cognitive health.

Details DOI

AAAI Conference 2025 Conference Paper

What Is a Good Question? Assessing Question Quality via Meta-Fact Checking

Bo Zhang
Jianghua Zhu
Chaozhuo Li
Hao Yu
Li Kong
Zhan Wang
Dezhuang Miao
Xiaoming Zhang

Knowledge-based questions are typically employed to evaluate LLM's knowledge boundaries; meanwhile, numerous studies focus on question generation as a means to enhance the capabilities of both models and individuals. However, there is a lack of in-depth exploration about what constitutes a good question from the perspective of knowledge cognition. This paper proposes aligning the complete knowledge underlying questions with educational criteria effectively employed in physics courses, thereby developing novel knowledge-intensive metrics of question quality. To this end, we propose Meta-Fact Checking (MFC), which transforms questions into knowledge graph (KG) triples utilizing LLMs through few-shot prompting, thereby quantifying question quality based on the patterns observed within these triples. MFC introduces a novel interaction mechanism for KGs that communicates meta-facts, illustrating the types of knowledge that KGs can offer to the LLM for reasoning questions, rather than relying solely on the original triples. This strategy ensures that MFC remains unaffected by unexplored triples that LLM has not yet encountered within KGs compared to the retrieve-while-reasoning routine. Experiments across multiple datasets and LLMs demonstrate that MFC significantly improves the accuracy and efficiency of both question answering and assessing. This research marks a pioneering effort to automate the evaluation of question quality based on cognitive capabilities.

PDF Details DOI

IJCAI Conference 2022 Conference Paper

Deexaggeration

Li Kong
Chuanyi Li
Vincent Ng

We introduce a new task in hyperbole processing, deexaggeration, which concerns the recovery of the meaning of what is being exaggerated in a hyperbolic sentence in the form of a structured representation. In this paper, we lay the groundwork for the computational study of understanding hyperbole by (1) defining a structured representation to encode what is being exaggerated in a hyperbole in a non-hyperbolic manner, (2) annotating the hyperbolic sentences in two existing datasets, HYPO and HYPO-cn, using this structured representation, (3) conducting an empirical analysis of our annotated corpora, and (4) presenting preliminary results on the deexaggeration task.

PDF Details DOI