Update README.md

README.md

@@ -1,36 +1,41 @@
----
-title: "Slipstream: Semantic Quantization for Multi-Agent Coordination"
-emoji: 📄
-colorFrom: blue
-colorTo: indigo
-sdk: gradio
-app_file: app.py
-pinned: false
-license: mit
-tags: ["semantic-quantization", "multi-agent-systems", "protocol-standards", "token-efficiency"]
----
-
-# Slipstream: Semantic Quantization for Efficient Multi-Agent Coordination
-
-This Space was generated from a research paper PDF.
-
-## What you can do here
-
-- **Live Quantizer**: Type messy natural language and watch it get quantized to a UCR anchor (the core demo!)
-- **Start here**: guided entry points (summary / limitations / thread)
-- **Gallery**: extracted figures or page previews
-- **Chat**: ask questions about the paper
-- **Share Kit**: generate a tweet thread / talk outline / FAQ
-- **Model Playground**: chat with a referenced HF model (requires `HF_TOKEN`)
-
-## Optional secrets
-
-If you add these as Space secrets, Chat + Share Kit become generative:
-
-- `HF_TOKEN`: Hugging Face token (read access is sufficient for inference; write is **not** needed at runtime)
-- `PAPER_LLM_MODEL`: e.g. `meta-llama/Meta-Llama-3-8B-Instruct` (or any chat-completion capable model)
-
-## Build provenance
-
-- Source PDF: `slipstream-paper.pdf`
-- Extracted pages: 7
+---
+title: 'Slipstream: Semantic Quantization for Multi-Agent Coordination'
+emoji: 📄
+colorFrom: blue
+colorTo: indigo
+sdk: gradio
+app_file: app.py
+pinned: false
+license: mit
+tags:
+- semantic-quantization
+- multi-agent-systems
+- protocol-standards
+- token-efficiency
+sdk_version: 6.5.1
+---
+
+# Slipstream: Semantic Quantization for Efficient Multi-Agent Coordination
+
+This Space was generated from a research paper PDF.
+
+## What you can do here
+
+- **Live Quantizer**: Type messy natural language and watch it get quantized to a UCR anchor (the core demo!)
+- **Start here**: guided entry points (summary / limitations / thread)
+- **Gallery**: extracted figures or page previews
+- **Chat**: ask questions about the paper
+- **Share Kit**: generate a tweet thread / talk outline / FAQ
+- **Model Playground**: chat with a referenced HF model (requires `HF_TOKEN`)
+
+## Optional secrets
+
+If you add these as Space secrets, Chat + Share Kit become generative:
+
+- `HF_TOKEN`: Hugging Face token (read access is sufficient for inference; write is **not** needed at runtime)
+- `PAPER_LLM_MODEL`: e.g. `meta-llama/Meta-Llama-3-8B-Instruct` (or any chat-completion capable model)
+
+## Build provenance
+
+- Source PDF: `slipstream-paper.pdf`
+- Extracted pages: 7