Sync model card title and 354-of-4,824-step Gemma 4 AR budget note

README.md

@@ -15,7 +15,7 @@ model-index:
 
 # DiffusionGemma 2: Latent Code-Graph Adapters & Checkpoints
 
-Trained two-seed LoRA adapters, pointwise readout heads, matched `google/gemma-4-26B-A4B-it` autoregressive LoRA weights, frozen `ImageDecoder`, and non-diffusion baseline checkpoints for ***Latent Roadmaps: DiffusionGemma State Transitions for Code-Graph Refactoring***.
+Trained two-seed LoRA adapters, pointwise readout heads, preliminary 354-of-4,824-step single-seed `google/gemma-4-26B-A4B-it` autoregressive LoRA weights, frozen `ImageDecoder`, and non-diffusion baseline checkpoints for ***Latent Roadmaps: Soft-Token Re-Feeding in DiffusionGemma for Code Graphs***.
 
 - **Author:** `bumblebeers` ([GitHub](https://github.com/bumblebeers))
 - **License:** Apache-2.0
@@ -30,8 +30,8 @@ Trained on 16x16 AST call-graph adjacency grids (3 training repositories: `fasta
 - **`expA_dg/cg_lora_dg_seed42_ep3_slim_ckpt.pt.gz`** & **`expA_dg/cg_lora_dg_seed1337_ep3_slim_ckpt.pt.gz`**: Two-seed LoRA DiffusionGemma (`r=64`, ~150M LoRA parameters + pointwise readout head `2816 -> 1152`, 4,824 steps).
 - **`expA_dg/cg_frozen_dg_ep3_slim_ckpt.pt.gz`**: Frozen DiffusionGemma control (trains only the pointwise readout head with frozen 26B base).
 
-### 2. Experiment A: Matched `Gemma 4 26B-A4B-it` Autoregressive LoRA (`expA_gemma4_ar/`)
-- **`expA_gemma4_ar/gemma4_ar_lora_r64_s42/`**: Multimodal `Gemma 4 26B-A4B-it` LoRA (`r=64`, 74.3M LM parameters, frozen vision tower) fine-tuned on the identical 16x16 RGB image + instruction inputs to emit 256 space-separated cell IDs.
+### 2. Experiment A: Preliminary Budget-Truncated `Gemma 4 26B-A4B-it` Autoregressive LoRA (`expA_gemma4_ar/`)
+- **`expA_gemma4_ar/gemma4_ar_lora_r64_s42/`**: Multimodal `Gemma 4 26B-A4B-it` LoRA (`r=64`, 74.3M LM parameters, frozen vision tower) fine-tuned on 16x16 RGB image + instruction inputs to emit 256 space-separated cell IDs (stopped at 354 of 4,824 planned optimizer steps = 7% of DG's schedule, single seed).
 
 ### 3. Experiment B: Continuous Symbol-Vector Refactoring (`expB_dg/`)
 - **`expB_dg/cg_expB_lora_dg_seed42_ep3_slim_ckpt.pt.gz`**: **Modified model: trained input adapter** (`LayerNorm(384) + Linear(384, 2816)`) + LoRA DiffusionGemma + `2816 -> 384` readout head. Raw `.npy` predictions for both seeds (`seed42` and `seed1337`, confirming `26.4±0.3%` seen / `27.6±1.0%` unseen and `17.9±0.1x` unchanged-slot drift) are hosted in [bumblebeer/dgg2-code-graph-benchmarks](https://huggingface.co/datasets/bumblebeer/dgg2-code-graph-benchmarks).