shaswatamitra/falcon-yara-bi-all-mpnet-base-v2

FALCON bi-encoder — YARA / all-mpnet-base-v2

Contrastive encoder fine-tuned to map CTI text and YARA rules into a shared embedding space. Backbone: sentence-transformers/all-mpnet-base-v2.

Test-set metrics

split	recall@1	F1	threshold	diag mean	off-diag mean
pretrained	0.7325	0.2365	0.6850	0.7315	0.4856
run_0	0.9509	0.9335	0.7089	0.9632	0.1041
run_1	0.9509	0.9300	0.7008	0.9628	0.0473
run_2	0.9509	0.9394	0.7046	0.9559	0.0670
run_3	0.9509	0.9326	0.7006	0.9470	0.0347
run_4	0.9509	0.9420	0.7107	0.9754	0.0106

Training

Symmetric InfoNCE / NT-Xent over in-batch negatives. Best checkpoint selected by validation loss.

• Run 0 — batch=16, epochs=5, lr=2e-05, schedule=constant, T=0.05
• Run 1 — batch=50, epochs=10, lr=2e-05, schedule=constant, T=0.05
• Run 2 — batch=70, epochs=30, lr=2e-05, schedule=constant, T=0.05
• Run 3 — batch=128, epochs=30, lr=5e-05, schedule=warmup_cosine, T=0.05
• Run 4 — batch=70, epochs=50, lr=2e-05, schedule=constant, T=0.07

Loading

from transformers import AutoModel, AutoTokenizer
tok   = AutoTokenizer.from_pretrained("shaswatamitra/falcon-yara-bi-all-mpnet-base-v2")
model = AutoModel.from_pretrained("shaswatamitra/falcon-yara-bi-all-mpnet-base-v2")

Citation

@article{mitra2025falcon,
  title={FALCON: Autonomous Cyber Threat Intelligence Mining with LLMs for IDS Rule Generation},
  author={Mitra, Shaswata and Bazarov, Azim and Duclos, Martin and Mittal, Sudip and Piplai, Aritran and Rahman, Md Rayhanur and Zieglar, Edward and Rahimi, Shahram},
  journal={arXiv preprint arXiv:2508.18684},
  year={2025}
}