MindForge pipeline trains small LLMs for full-cycle software engineering

Researchers have developed MindForge, a novel pipeline designed to train smaller language models in comprehensive software engineering tasks. This system converts open-source command-line programs into source-free training environments, providing only compiled executables and documentation. By fine-tuning the Qwen3.6-27B model on synthesized data generated by GLM-5.2, the researchers achieved a significant improvement in its ability to generate complete programs, reaching performance levels comparable to much larger frontier models. AI

IMPACT Enhances the capabilities of smaller language models in complex software generation, potentially democratizing advanced coding assistance.

RANK_REASON The cluster describes a new research paper detailing a novel method for training language models on software engineering tasks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Deepsweg
• FeatBench
• GLM-5.2
• NL2Repo-Bench
• ProgramBench
• Qwen3.6-27B
• RepoZero-C2Rust
• SWE-bench Multilingual
• SWE Bench Pro
• SWE-bench Verified

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →