EF Poles Classifier

Fine-tuned Qwen3-0.6B via GRPO for classifying EU development projects to Expertise France thematic poles and transversal axes.

Input: Project description + DAC codes + country
Output: List of EF poles (1–22) + transversal axes (1–6)
Reward: Composite — 0.2 × clean termination + 0.8 × semantic score, where semantic score = 0.1 (format bonus) + 0.9 × Jaccard similarity over a unified 28-class space (poles 1–22, transversal axes encoded as 23–28)


Training Data

The model was trained on a synthetic dataset (JZSG/ef_training_datasets) generated with Gemini 3.0 Flash from 1 442 original labeled examples, using a four-phase pipeline:

  1. Rebalancing of under-represented poles
  2. Variations on existing examples
  3. Country diversification
  4. Edge cases and ambiguous assignments

The synthetic dataset was split into train / test sets. Training was done on the train split; all evaluation figures below are on the held-out test split.

Split Examples
Train 10,295
Test 1,144

Training Setup

Parameter Value
Base model Qwen3-0.6B
Method GRPO
Infrastructure Jean Zay (IDRIS)
Temperature 1.0

Training pipelines and evaluation scripts: Pleias/EF_training (private).

Downloads last month
1
Safetensors
Model size
0.6B params
Tensor type
F32
·
Video Preview
loading

Model tree for JZSG/ef_qwen3_0.6B_poles

Finetuned
Qwen/Qwen3-0.6B
Finetuned
(799)
this model

Dataset used to train JZSG/ef_qwen3_0.6B_poles