Contrastive Learning on AIBussin — AI applications, systems and books

Hard Negatives

Mon, 07 Sep 2026 11:20:00 +0000

Part III — Retrieval Is an Experiment

Same model, four negative sets, four verdicts

MEASURED on RELATE v0.1, Wave 1 row 1.8 — artifact experiments/embeddings-from-first-principles/wave1/artifacts/margin-collapse.json. Model all-mpnet-base-v2, unchanged across every row below.

negative-selection rule                   mean margin   selected-positive win rate
same-domain random, up to 5 per query        0.4738              1.0000
in-model selector, as implemented            0.2094              0.9814
structured perturbations                     0.0932              0.9331
BM25 lexical top-5                           0.0595              0.7621

The model did not change. The 269-query set did not change. The reference-positive rule did not change. Only the rule that picked which passages the positive had to beat changed — and the mean margin ranged from 0.4738 to 0.0595, a nearly eightfold difference between the widest and narrowest measured conditions.

Preference Rankers — Learning Which Answer Is Better

Sat, 22 Aug 2026 01:18:00 +0100

Preference Rankers — Learning Which Answer Is Better

So far in Models From First Principles, we have mostly trained models by telling them what the answer should be.

MR.Q took a context and a response and produced a number:

context + response
        ↓
      model
        ↓
      0.82

That number might mean quality.

Or usefulness.

Or reward.

But there is an awkward question hiding underneath that architecture:

Where did the 0.82 come from?

Everything is a Trace: Stephanie Enters Full Reflective Mode

Sun, 03 Aug 2025 22:03:10 +0100

🔧 Summary

In our last post, Layers of Thought: Smarter Reasoning with the Hierarchical Reasoning Model, we introduced a new epistemic lens a way to evaluate not just final answers, but the entire sequence of reasoning steps that led to them. We realized we could apply this way of seeing to every action in our system not just answers, but inferences, lookups, scorings, decisions, and even model selections. This post shows how we’re doing exactly that.