Update README.md

README.md

@@ -19,20 +19,6 @@ This repository releases **LoRA + expanded input-projection weights** for **vide
 
 > **Disclaimer:** This is a **supplementary** experiment from the CLEAR project. The main paper results use **Wan2.1-Control**; this CogVideoX-2b variant is **not** expected to match that baseline. It is shared for **reproducibility and comparison**.
 
-## What is included
-
-- **Checkpoint:** `cogvideox_2b_CLEAR_lora_checkpoint.pt`  
-  Contains:
-  - `lora_state_dict` — LoRA on attention (`to_q`, `to_k`, `to_v`, `to_out.0`)
-  - `proj_state_dict` — expanded `patch_embed.proj` (16→32 input channels for conditioning)
-  - `step` — training step metadata
-
-- **Training / inference code:** see linked project or bundled scripts (if you attach the `release/` archive to the repo).
-
-## What is **not** included
-
-- **Training data** (not released).
-- **Base model weights** — load from `zai-org/CogVideoX-2b`.
 
 ## Architecture change (high level)