Middle Code Prediction: Enhancing Code Generation for Uncommon Programming Languages in Robotics

Zixi Jia, Hongbin Gao, Hexiao Li
Proceedings of the 16th Asian Conference on Machine Learning, PMLR 260:1112-1127, 2025.

Abstract

Generating executable code through natural language instructions to drive robotic movements is considered a crucial step towards achieving embodied intelligence. However, in the robotics domain, the scarcity of programming language data often necessitates manually encapsulating high-level APIs to enable Large Language Models(LLMs) to predict code correctly, which is time-consuming and incomplete. Therefore, this paper proposes a three-stage Middle Code Prediction(MCP) scheme, by injecting appropriate prompts at different stages, the LLMs can shift towards predicting middle code that it understands more easily. This middle code can then be converted into the final code through specific scripts, accomplishing the task of generating code in uncommon programming languages automatically and without the need for manually encapsulating high-level APIs. We tested our approach on Hospital Item Transport Dataset(HITD) and found that MCP could improve the mean accuracy of various baseline models to varying degrees, with an overall increase of 31%, while also enhancing the noise resistance of fine-tuned models. We conducted real-world experiments on industrial robotic arms, verifying the feasibility of MCP in scenarios with no API and partial API encapsulation. The method proposed in this paper provides a guideline for code generation in uncommon programming languages within the context of LLMs. Our experimental dataset is available at https://github.com/Ghbbbbb/MCP.

Cite this Paper

BibTeX
@InProceedings{pmlr-v260-jia25c, title = {{Middle Code Prediction}: {E}nhancing Code Generation for Uncommon Programming Languages in Robotics}, author = {Jia, Zixi and Gao, Hongbin and Li, Hexiao}, booktitle = {Proceedings of the 16th Asian Conference on Machine Learning}, pages = {1112--1127}, year = {2025}, editor = {Nguyen, Vu and Lin, Hsuan-Tien}, volume = {260}, series = {Proceedings of Machine Learning Research}, month = {05--08 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v260/main/assets/jia25c/jia25c.pdf}, url = {https://proceedings.mlr.press/v260/jia25c.html}, abstract = {Generating executable code through natural language instructions to drive robotic movements is considered a crucial step towards achieving embodied intelligence. However, in the robotics domain, the scarcity of programming language data often necessitates manually encapsulating high-level APIs to enable Large Language Models(LLMs) to predict code correctly, which is time-consuming and incomplete. Therefore, this paper proposes a three-stage Middle Code Prediction(MCP) scheme, by injecting appropriate prompts at different stages, the LLMs can shift towards predicting middle code that it understands more easily. This middle code can then be converted into the final code through specific scripts, accomplishing the task of generating code in uncommon programming languages automatically and without the need for manually encapsulating high-level APIs. We tested our approach on Hospital Item Transport Dataset(HITD) and found that MCP could improve the mean accuracy of various baseline models to varying degrees, with an overall increase of 31%, while also enhancing the noise resistance of fine-tuned models. We conducted real-world experiments on industrial robotic arms, verifying the feasibility of MCP in scenarios with no API and partial API encapsulation. The method proposed in this paper provides a guideline for code generation in uncommon programming languages within the context of LLMs. Our experimental dataset is available at https://github.com/Ghbbbbb/MCP.} }
Endnote
%0 Conference Paper %T Middle Code Prediction: Enhancing Code Generation for Uncommon Programming Languages in Robotics %A Zixi Jia %A Hongbin Gao %A Hexiao Li %B Proceedings of the 16th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Vu Nguyen %E Hsuan-Tien Lin %F pmlr-v260-jia25c %I PMLR %P 1112--1127 %U https://proceedings.mlr.press/v260/jia25c.html %V 260 %X Generating executable code through natural language instructions to drive robotic movements is considered a crucial step towards achieving embodied intelligence. However, in the robotics domain, the scarcity of programming language data often necessitates manually encapsulating high-level APIs to enable Large Language Models(LLMs) to predict code correctly, which is time-consuming and incomplete. Therefore, this paper proposes a three-stage Middle Code Prediction(MCP) scheme, by injecting appropriate prompts at different stages, the LLMs can shift towards predicting middle code that it understands more easily. This middle code can then be converted into the final code through specific scripts, accomplishing the task of generating code in uncommon programming languages automatically and without the need for manually encapsulating high-level APIs. We tested our approach on Hospital Item Transport Dataset(HITD) and found that MCP could improve the mean accuracy of various baseline models to varying degrees, with an overall increase of 31%, while also enhancing the noise resistance of fine-tuned models. We conducted real-world experiments on industrial robotic arms, verifying the feasibility of MCP in scenarios with no API and partial API encapsulation. The method proposed in this paper provides a guideline for code generation in uncommon programming languages within the context of LLMs. Our experimental dataset is available at https://github.com/Ghbbbbb/MCP.
APA
Jia, Z., Gao, H. & Li, H.. (2025). Middle Code Prediction: Enhancing Code Generation for Uncommon Programming Languages in Robotics. Proceedings of the 16th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 260:1112-1127 Available from https://proceedings.mlr.press/v260/jia25c.html.

Related Material