Model card: Decision Index 0.3 public suite scores (63.47)

README.md

@@ -13,9 +13,12 @@ base_model:
 
 # MATILDA JEV v1.5 by Maincode
 
+> All scores on this card are from the **Decision Index 0.3** public suite (the public dataset of the current
+> leaderboard edition).
+
 A merged BF16 JEV decision model with a 255-option readout. This is the
 highest-scoring checkpoint in the completed MATILDA JEV experiments as of
-9 October 2026: **Decision Index 62.76**.
+9 October 2026: **Decision Index 63.47** (0.3 public suite).
 
 The model accepts a state and one or more decision questions. It supports
 `choice`, `noul` (probability of true), and ordinal `score`. Inference is one
@@ -69,20 +72,23 @@ python /path/to/downloaded/model/predict.py --device cuda:0 < requests.jsonl
 
 ## Evaluation
 
-Decision Index edition 0.2.1: **150,317 successful requests across 44 benchmarks**.
-Saved predictions were independently rescored with identical results. The full
-results are in [`scores.json`](scores.json).
+Decision Index 0.3 public suite: **140,178 of 140,178 scored requests answered across 37 benchmarks**
+(kit [`62d2f51`](https://github.com/apolinario/decision-index/commit/62d2f51)).
+Saved predictions were independently rescored with identical results. The full run, including every
+prediction, is in [Maincode/matilda-jev-decision-index](https://huggingface.co/datasets/Maincode/matilda-jev-decision-index)
+under `runs/matilda-jev-v1.5/`. [`scores.json`](scores.json) in this repository is the earlier 0.2.1 run
+(62.76), kept for reference.
 
 | Metric | Score |
 |---|---:|
-| Decision Index | 62.76 |
-| Raw | 71.64 |
-| Breadth | 61.51 |
-| Knowledge & Reasoning | 47.75 |
-| Language Understanding | 70.31 |
-| Retrieval & Classification | 65.97 |
+| Decision Index | 63.47 |
+| Raw | 72.62 |
+| Breadth | 62.45 |
+| Knowledge & Reasoning | 49.83 |
+| Language Understanding | 69.91 |
+| Retrieval & Classification | 65.96 |
 | Tools & Automation | 80.84 |
-| Arts & Human Taste | 42.53 |
+| Arts & Human Taste | 45.41 |
 
 Area scores are chance-corrected skill multiplied by 100. These results are
 diagnostic: earlier training in this model's lineage used benchmark-related