machine learning

12 posts tagged machine-learning.

Stephen Collins

When You Don't Have Enough History to Train

Every new repo starts with the same problem: no defect labels, no training signal, no basis for a risk ranking beyond 'how often does this file change?'

Jun 24, 2026 Read →
Stephen Collins

The Signal That Evaporated (And the One That Didn't)

Two signals looked nearly identical on paper. One collapsed under temporal holdout. The other survived — and the difference tells you something important about how to design defect prediction features…

Jun 24, 2026 Read →
Stephen Collins

Spearman ρ Looked Fine. P@10 Was Zero.

Spearman ρ can be positive and meaningful while the top-ranked files contain zero future bugs. Here's how I found that out, why it happens, and how to catch it before it reaches a user.

Jun 24, 2026 Read →
Stephen Collins

Tabular Beats LLM, and That's the Right Answer

On most codebases, a small XGBoost model trained on git history features outperforms a fine-tuned 7B LLM at defect risk ranking — not because the LLM is bad, but because the data is inherently tabular.

Jun 24, 2026 Read →
Stephen Collins

The Coupling Signal That Kept Inverting Itself

I tried three formulations of co-change coupling for defect risk ranking. Raw co-occurrence inverted on dense codebases. Mutual information made it worse.

Jun 10, 2026 Read →
Stephen Collins

Your Transformer Featurizer Only Helps When Your History Is Broken

Transformer embeddings on source code improved defect ranking on exactly the repos where git history had already failed — and added near-zero or negative signal everywhere else. Here's what that conditional looks like in practice.

Jun 10, 2026 Read →
Stephen Collins

The Defect Prediction Trap

I fine-tuned a model that scored Spearman ρ=+0.678 on defect risk. Then I zeroed out the git history fields and it dropped to −0.142. The model wasn't reading code. It was relaying commit metadata back as a prediction.

Jun 5, 2026 Read →
Stephen Collins

The Metric That Hides Failure

Spearman ρ measures average rank agreement across all files. Precision@K measures whether the files engineers actually review are in the right place. On mature, stable codebases, a ranker can show positive ρ while returning zero bugs in its top 10.

Jun 5, 2026 Read →
Stephen Collins

Which LLM Vendor Wins at Defect Risk Prediction?

When evaluating open-weight models for code analysis, the vendor comparison you run in 2024 might give you the wrong answer in 2025.

Jun 5, 2026 Read →
Stephen Collins

When Your Labels Lie: File-Level Bug Attribution Inflates Defect Rates

Most defect prediction pipelines share a quiet assumption: if a file was touched in a bug-fix commit, every function in that file was 'buggy' — and that assumption silently inflates defect labels in ways that undermine fine-tuning before it starts.

May 31, 2026 Read →
Stephen Collins

How to Know If Fine-Tuning Will Help Before You Run It

Before you spend compute on per-repo fine-tuning, one scalar from the codebase's bug-fix history can tell you whether it's worth running. Here's how it works and what it can't predict.

May 25, 2026 Read →
Stephen Collins

Can a Fine-Tuned LLM Learn Which Code Is Risky?

I spent a few weeks running a controlled experiment: can a small language model, fine-tuned on a specific codebase's history, learn to rank code by defect risk better than a hand-tuned heuristic?

May 16, 2026 Read →

Other Tags