Arrow Research search

Author name cluster

Zhen Cai

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.

2 papers
1 author row

Possible papers

2

EAAI Journal 2026 Journal Article

Deep learning-based on-site identification and volume measurement of bulk material in construction industry

  • Zhen Cai
  • Ghaith Allah Chebil
  • Yuan Tan
  • Stephan Kessler
  • Johannes Fottner

Bulk materials are important raw construction materials, the adequate and precise supply of which enables a smooth construction process. Two conventional techniques for controlling the quantity of bulk materials on-site along the supply chain are: 1) estimation based on the cone-shape of the material pile, but the accuracy is low; 2) calculation by using bulk density and weighing stations, which are not available in all facilities. To address this issue, we propose a novel hybrid camera-based method combining a red-green-blue (RGB) camera and a light detection and ranging (LiDAR) sensor for automatic material type identification and volume measurement. The data from two-dimensional pictures and three-dimensional point cloud were segmented and extracted with two Deep Learning models: “You Only Look Once“ (YOLO) v5 and PointNet++for the identification of material type and volume measurement. This novel hybrid camera-based method was developed as an Industry 4. 0 solution to enable the automatic and accurate evaluation of the volume of bulk materials. With a precision of 81. 6 % in object recognition and a volume estimation deviation of less than 8 %, it provides a reliable and efficient alternative to conventional, labour-intensive measurement techniques.

Details DOI

AAAI Conference 2025 Conference Paper

LightPROF: A Lightweight Reasoning Framework for Large Language Model on Knowledge Graph

  • Tu Ao
  • Yanhua Yu
  • Yuling Wang
  • Yang Deng
  • Zirui Guo
  • Liang Pang
  • Pinghui Wang
  • Tat-Seng Chua

Large Language Models (LLMs) have impressive capabilities in text understanding and zero-shot reasoning. However, delays in knowledge updates may cause them to reason incorrectly or produce harmful results. Knowledge Graphs (KGs) provide rich and reliable contextual information for the reasoning process of LLMs by structurally organizing and connecting a wide range of entities and relations. Existing KG-based LLM reasoning methods only inject KGs' knowledge into prompts in a textual form, ignoring its structural information. Moreover, they mostly rely on close-source models or open-source models with large parameters, which poses challenges to high resource consumption. To address this, we propose a novel Lightweight and efficient Prompt learning-ReasOning Framework for KGQA (LightPROF), which leverages the full potential of LLMs to tackle complex reasoning tasks in a parameter-efficient manner. Specifically, LightPROF follows a “Retrieve-Embed-Reason” process, first accurately, and stably retrieving the corresponding reasoning graph from the KG through retrieval module. Next, through a Transformer-based Knowledge Adapter, it finely extracts and integrates factual and structural information from the KG, then maps this information to the LLM’s token embedding space, creating an LLM-friendly prompt to be used by the LLM for the final reasoning. Additionally, LightPROF only requires training Knowledge Adapter and can be compatible with any open-source LLM. Extensive experiments on two public KGQA benchmarks demonstrate that LightPROF achieves superior performance with small-scale LLMs. Furthermore, LightPROF shows significant advantages in terms of input token count and reasoning time.

PDF Details DOI
v2026.09.13