Feihu Ma

Prof. Dr. Feihu Ma

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 August 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
FMFeihu Ma
Feihu Ma
SWShuhan Wang
Shuhan Wang
TXTianchang Xie
Tianchang Xie
CSCuiyu Sun
Cuiyu Sun
In the current era, the government consistently emphasizes the pursuit of high-quality development, as evidenced by the ongoing increase in the tertiary industry’s GDP share. As a crucial component of the modern service sector, logistics plays a pivotal role in determining the operational efficiency and overall quality of the industrial economy. This study focuses on constructing a Chongqing logistics express demand prediction index system. It employs an improved BP neural network model to forecast the logistics express demand for Chongqing over the next five years. Given the limited express demand data sequence and the normalized characteristics of the data, the selected training method is the Bayesian regularization approach, with the LeCun Tanh function serving as the hidden layer activation function. Additionally, a genetic algorithm is designed to optimize the initial weights and thresholds of the BP neural network, thereby enhancing prediction accuracy and reducing the number of iterations. The experimental results of the improved GA-BP network are analyzed and compared, demonstrating that the improved BP neural network, utilizing GA optimization, can more reliably and accurately predict regional logistics express demand. According to the findings, the forecast indicates that the logistics express demand for Chongqing in 2026 will be 2,171,642,700 items.

30 September 2024 in IEEE Access
Verified
Journal Article
Open access
FMFeihu Ma
Feihu Ma
薛梁薛辰 梁
薛辰 梁
MTMeiling Tao
Meiling Tao
As large language models (LLMs) improve their capabilities in handling complex tasks, the issues of computational cost and efficiency due to long prompts are becoming increasingly prominent. To accelerate model inference and reduce costs, we propose an innovative prompt compression framework called LinguaShrink. Inspired by the observation that LLM performance depends on the density and position of key information in the input prompts, LinguaShrink leverages psycholinguistic principles and the Ebbinghaus memory curve to achieve task-agnostic prompt compression. This effectively reduces prompt length while preserving essential information. We referred to the training method of OpenChat [1]. The framework introduces part-of-speech priority compression and data distillation techniques, using smaller models to learn compression targets and employing a KL-regularized reinforcement learning strategy for training. Additionally, we adopt a chunk-based compression algorithm to achieve adjustable compression rates. We evaluate our method on multiple datasets, including LongBench, ZeroScrolls, Arxiv Articles, and a newly constructed novel test set. Experimental results show that LinguaShrink maintains semantic similarity while achieving up to 26 times compression. Compared to existing prompt compression methods, LinguaShrink improves end-to-end latency by 1.43 times.

24 June 2025 in IEEE Access
Verified
Journal Article
Open access
FMFeihu Ma
Feihu Ma
薛梁薛辰 梁
薛辰 梁
JLJianing Long
Jianing Long
MTMeiling Tao
Meiling Tao
Recent advancements in large language models (LLMs) have unlocked the potential for generating high-quality game commentary. However, producing insightful and engaging commentary for complex games with incomplete information remains a significant challenge. In this paper, we introduce a novel commentary method that combine Reinforcement Learning (RL) and LLMs, tailored specifically for the Chinese card game Guandan. Our system leverages RL to generate intricate card-playing scenarios and employs LLMs to generate corresponding commentary text, effectively emulating the strategic analysis and narrative prowess of professional commentators. The framework comprises a state commentary guide, a Theory of Mind (ToM)-based strategy analyzer, and a style retrieval module, which seamlessly collaborate to deliver detailed and context-relevant game commentary in the Chinese language environment. We empower LLMs with ToM capabilities and refine both retrieval and information filtering mechanisms. This facilitates the generation of personalized commentary content. The semantic similarity of our model far exceeds that of other models by close to 80%. Our experimental results showcase the substantial enhancement in performance achieved by the proposed commentary framework when applied to open-source LLMs, surpassing the performance of GPT-4 across multiple evaluation metrics.