Modelpublic

aryasuneesh-quilr/intent-crossencoder-minilm-L12-v2

sourceHugging Faceapache-2.0updated 7mo agoView on Hugging Face

0likes62downloads

Model Card

intent-crossencoder-minilm-L12-v2

Cross-encoder reranker for enterprise intent detection (DLP / security). Fine-tuned from cross-encoder/ms-marco-MiniLM-L12-v2 on a synthetic intent-detection dataset.

Intended use

Binary classification: given a (user_input, intent_description) pair, predict whether the user input matches the intent. Designed as Stage-2 in a cascading firewall: Stage-1 (fast heuristic) → this model (reranker) → Stage-2 LLM (Qwen).

Input format

user_input [SEP] intent_description

Performance (held-out test set, threshold=0.3)

Metric	Value
Recall	0.9966
Precision	0.9758
F1	0.9861
AUC-ROC	0.9949
PR-AUC	0.9899
Best threshold (F1-optimal)	0.9655 → F1=0.9888

Training config

Parameter	Value
Base model	`cross-encoder/ms-marco-MiniLM-L12-v2`
Batch size	64
Grad accum steps	1
Effective batch	64
Learning rate	2e-05
Label smoothing	0.05
Warmup ratio	0.06
Max sequence length	256
Early stopping	recall@0.3 (patience=3)
Epochs trained	7
Training time	3.8 min

Inference snippet

python

import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

model_id  = "aryasuneesh-quilr/intent-crossencoder-minilm-L12-v2"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model     = AutoModelForSequenceClassification.from_pretrained(model_id)
model.eval()

def score(user_input: str, intent_description: str) -> float:
    pair = f"{user_input} [SEP] {intent_description}"
    enc  = tokenizer(pair, return_tensors="pt", truncation=True, max_length=256)
    with torch.no_grad():
        logits = model(**enc).logits
    return torch.softmax(logits, dim=1)[0, 1].item()   # P(match)

# Example
s = score(
    "Our AWS_SECRET_ACCESS_KEY was found in a public repo",
    "Identify exposure of authentication credentials or API keys"
)
print(f"Match probability: {s:.4f}")   # use threshold 0.9655 for best F1

Files in this repo

File	Description
`model.safetensors`	HF-native weights
`best_model.pt`	Raw PyTorch state_dict (for resuming training)
`training_config.json`	Full hyperparameter record
`metrics/`	Per-epoch + test-set evaluation CSVs

Generated 2026-02-23 08:36 UTC by ablation_reranker_training.py