PAFT Artifact

This repository contains the implementation and public artifact for PAFT (Preservation-Aware Fine-Tuning), a fine-tuning workflow for minimal-edit program repair.

What Is Included

• Core training and inference code: SingleTrainWithLCS.py, merge_adapter.py, defects4j.py, test_d4j.py, inference_java.py, calc_java.py, and stats_diff_java.py.
• Benchmark-facing assets and scripts for Defects4J and HumanEval-Java: defects4j/, evalrepair-java/, and table*.sh.
• TSE revision evidence and audit summaries under analysis_outputs/.
• Reproduction and audit utilities under scripts/.

Large model weights, tokenizer snapshots, generated logs, and full raw result directories are intentionally excluded from Git. The scripts expect those assets to be downloaded or regenerated locally.

Key Reproduction Commands

# Original DS-Coder-6.7B training/evaluation pipeline
./pipeline_deepseek-6.7b.sh

# Reuse an existing trained model and run downstream stages
./pipeline_deepseek-6.7b.sh --skip-training

# Validate Defects4J generations and compute repair metrics
python test_d4j.py -m <model-name> -n 10

# Re-evaluate HumanEval-Java outputs
python calc_java.py <model-name> rejudge
python stats_diff_java.py -m <model-name>

TSE Revision Evidence

The fixed-seed evidence used for the journal revision is summarized in:

• analysis_outputs/tse_fixed_seed_manifest.md
• analysis_outputs/tse_current_evidence_summary.md
• analysis_outputs/tse_evidence_artifact_check.md

The local paper-readiness gate, which expects the separate tse-paper/ manuscript checkout to exist next to the artifact files, is:

scripts/check_tse_ready.sh

Environment Notes

Training requires a CUDA GPU with at least 24 GB VRAM, Python 3.8+, PyTorch, Transformers, PEFT, bitsandbytes, and benchmark dependencies such as Defects4J. Set model and dataset paths explicitly in scripts or command-line arguments; do not commit local credentials, API keys, checkpoints, or generated model files.