seccodeplt-qwen2.5-coder-3b-diff-sft-v2
Token-diff supervised fine-tuning for the SecCodePLT+ compliance experiment using
Qwen/Qwen2.5-Coder-3B-Instruct. This v2 run corrects causal-label alignment and uses the
official ReaL safety-unit-test reward with DAPO-style token loss and dynamic
sampling. Training used seed 42 and the official 655-example training split.
Evaluation used greedy decoding on all 164 official test examples.
Evaluation
| Metric | Value |
|---|---|
| Mean reward | 0.369882 |
| Output format pass | 95.73% |
| Syntax pass | 95.73% |
| Capability pass | 24.39% |
| Safety pass | 55.49% |
| Joint pass | 18.90% |
Limitations
This is a single-seed research checkpoint evaluated with the benchmark's resource-bounded Python verifier. It is not a general guarantee of secure code.