Update README.md

README.md

@@ -19,4 +19,9 @@ This repository hosts pre-trained model checkpoints for cross-modal video-text u
 | mamba_mobileclip_ckpt.pt | 500 MB | Mamba trained to aggregate MobileCLIP embeddings. Checkpoint 6900, no FiLM |
 | internvideo2_clip.pt    | 5.55 MB | CLIP component of **InternVideo2-B14** |
 | internvideo2_vision.pt  | 205 MB  | Vision encoder backbone for **InternVideo2-B14** |
-| mobileclip_blt.pt       | 599 MB  | Lightweight **MobileCLIP** variant (BLT) |
+| mobileclip_blt.pt       | 599 MB  | Lightweight **MobileCLIP** variant (BLT) |
+
+| spfs_r64 files | Size | Description |
+| ------------------------|---------|---------------------|
+| /confidence_head.pt       | 144 MB  | Confidence head for SPFS, saved at step 14,000 (Epoch 2) |
+| /predictor_head.pt       | 3.04 MB  | Predictor head for SPFS, saved at the end of Epoch 1 |