docs: self-citation, update-clipping recipe note, long-context limitation

README.md

@@ -84,7 +84,7 @@ Also mirrored on ModelScope:
 - Data: [JessieWei/GLM-5.2-FP8-nemotron-codealpaca](https://huggingface.co/datasets/JessieWei/GLM-5.2-FP8-nemotron-codealpaca)
   — Nemotron-Post-Training-v2 + CodeAlpaca, ~800K samples (paper-specified recipe),
   all responses regenerated by GLM-5.2-FP8 (non-thinking mode), max_length 3072
-- Recipe: 6 epochs, AdamW, lr 6e-4 cosine (4% warmup), grad-clip 1.0, num_anchors 512,
+- Recipe: 6 epochs, AdamW with update clipping (StableAdamW-style), lr 6e-4 cosine (4% warmup), grad-clip 1.0, num_anchors 512,
   loss_decay_gamma 7, pure cross-entropy (standard DFlash loss)
 - Framework: SpecForge (offline hidden-state pipeline), FSDP2
 
@@ -92,6 +92,8 @@ Also mirrored on ModelScope:
 
 - Trained on non-thinking-mode regenerated data; speedup under thinking-mode inference
   has not been evaluated yet.
+- Trained with `max_length` 3072: accept length degrades on very long prompts. For
+  long-context workloads, benchmark against the built-in MTP before choosing.
 - Training corpus is English-dominant: acceptance length on Chinese-language
   workloads is lower (ceval AL 2.98 vs 4.4-8.0 on English benchmarks).
 
@@ -104,6 +106,19 @@ GLM-5.2 by Zhipu AI.
 
 ## Citation
 
+If you use this model, please cite:
+
+```bibtex
+@misc{ucloud2026glm52dflash,
+  title        = {GLM-5.2-FP8-DFlash: A DFlash Speculative-Decoding Drafter for GLM-5.2-FP8},
+  author       = {Wei, Xiaojun and {UCloud AILab}},
+  year         = {2026},
+  howpublished = {\url{https://huggingface.co/UCloud-org/GLM-5.2-FP8-DFlash}}
+}
+```
+
+This model is trained with the DFlash method — please also cite:
+
 ```bibtex
 @misc{chen2026dflash,
   title         = {DFlash: Block Diffusion for Flash Speculative Decoding},