Update README.md

README.md

@@ -7,7 +7,6 @@ language:
 - en
 pipeline_tag: text-generation
 library_name: transformers
-
 tags:
 - stablecoin
 - payments
@@ -21,7 +20,6 @@ tags:
 - web3
 - usdc
 - eip-3009
-- compliance
 - fraud-detection
 - financial-risk
 - qwen
@@ -29,80 +27,24 @@ tags:
 - text-generation
 ---
 
-Base Model: huggingface.co/sriram1983007/SRA-RiskGate-4B
-
-GGUF Repo: huggingface.co/sriram1983007/SRA-RiskGate-4B-Q4_K_M-GGUF
-
-### 🦙 Run Instantly with Ollama
-
-```bash
-ollama run sriram1983007/sra-riskgate
-
-# SRA-RiskGate-4B
-
-A small model for both ends of a stablecoin payment's life:
-
-- **Risk gate:** give it a payment (an x402 request, an EIP-3009 authorization, or an ERC-20 transfer), your
-  policy, and your verification tool results; it returns `approve` / `hold` / `reject` with reasons.
-- **Dispute adjudication:** give it a dispute with signed evidence and escrow/ledger state; it decides
-  what can actually happen across the settlement-finality line (void before release, arbiter refund from
-  escrow or from a merchant bond, voluntary refund, deny, escalate, or no enforceable remedy), with exact
-  amount, destination and an idempotency key for any refund. It never proposes reversing a final payment. Fine-tuned from `Qwen/Qwen3-4B-Instruct-2507` on [sriram1983007/sra-stablecoin-risk-bench](https://huggingface.co/datasets/sriram1983007/sra-stablecoin-risk-bench).
-
-# 🛡 SRA-RiskGate-4B
+# 🛡️️ SRA-RiskGate-4B
 
 [![Interactive Demo](https://img.shields.io/badge/Demo-Interactive%20Space-blue?style=for-the-badge&logo=huggingface)](https://sriram1983007-sra-riskgate-demo.static.hf.space/index.html)
 [![Dataset](https://img.shields.io/badge/Dataset-Risk%20Bench-green?style=for-the-badge&logo=huggingface)](https://huggingface.co/datasets/sriram1983007/sra-stablecoin-risk-bench)
+[![GGUF Quantized](https://img.shields.io/badge/GGUF-Q4__K__M-yellow?style=for-the-badge&logo=huggingface)](https://huggingface.co/sriram1983007/SRA-RiskGate-4B-Q4_K_M-GGUF)
+[![Ollama Registry](https://img.shields.io/badge/Ollama-Registry-black?style=for-the-badge&logo=ollama)](https://ollama.com/sriram1983007/sra-riskgate)
 
-**SRA-RiskGate-4B** is an autonomous risk scoring, compliance verification, and dispute resolution model fine-tuned on top of `Qwen/Qwen3-4B-Instruct-2507`.
-
-* **Overall SRA Score:** `0.9159`
-* **Risk Decision Accuracy:** `99.49%`
-* **Unsafe Approval Rate:** `0.47%` (Threshold: $\le 2.0\%$)
-* **Dispute Impossible Remedy Rate:** `0.19%` (Threshold: $\le 1.0\%$)
-* **JSON Schema Validity:** `100.0%`
-  
-## Results on the SRA benchmark (test split, 2,000 payments)
-
-| model | sra_score | risk gate score | unsafe approvals | dispute score | impossible remedies | wrongful refunds | refund detail acc. |
-|---|---|---|---|---|---|---|---|
-| Qwen/Qwen3-4B-Instruct-2507 (untrained) | 0.421 | 0.443 | 0.345 | 0.400 | 0.000 | 0.000 | 0.000 |
-| **SRA-RiskGate-4B** | 0.916 | 0.995 | 0.005 | 0.837 | 0.002 | 0.022 | 0.660 |
-
-`unsafe approvals`: payments that should have been held or rejected but were approved.
-`impossible remedies`: on final payments, reversals, escrow mechanisms, or forced refunds with no usable bond.
-`wrongful refunds`: refunds granted where the correct outcome was not a refund.
-Injection prompts in the test split use phrasings never seen in training.
-
-## How to use
+An autonomous risk scoring, compliance verification, and dispute adjudication model fine-tuned from `Qwen/Qwen3-4B-Instruct-2507` on [sra-stablecoin-risk-bench](https://huggingface.co/datasets/sriram1983007/sra-stablecoin-risk-bench).
 
-```python
-import json
-from transformers import pipeline
-gate = pipeline("text-generation", model="sriram1983007/SRA-RiskGate-4B", device_map="auto")
+A specialized 4B model designed for both ends of a stablecoin payment lifecycle:
+- **Risk Gate:** Evaluates incoming payment operations (x402 requests, EIP-3009 authorizations, or raw ERC-20 transfers) against policy constraints and verification tool results to return deterministic `approve`, `hold`, or `reject` verdicts with structured reasoning.
+- **Dispute Adjudication:** Ingests dispute evidence and escrow/ledger state to determine legally and technically viable remedies across settlement-finality boundaries (void before release, arbiter escrow refund, merchant bond claim, voluntary refund, deny, or escalate) complete with exact amount, destination address, and idempotency keys. It strictly avoids proposing impossible reversals on final ledger state.
 
-# see sra/prompt.py in the source repo
-messages = build_messages(now, policy, payload, tool_results)                  # risk gate
-# messages = build_dispute_messages(now, policy, case, tool_results)          # dispute
-out = gate(messages, max_new_tokens=512, do_sample=False)[0]["generated_text"][-1]["content"]
-verdict = json.loads(out)
-```
-
-The model expects **tool results** (signature verification, attestation checks, sanctions screening) in the
-prompt. It does not do cryptography and does not know any sanctions list. Always run live screening
-through a tool.
-
-## Intended use and limits
+---
 
-- A **triage and explanation layer** in front of human or rule-based controls for agent payments,
-  stablecoin operations and dispute desks. It is not a compliance program, not an arbiter, and not legal
-  advice. Dispute recommendations should be reviewed by a person before funds move.
-- Trained on synthetic data with templated explanations; expect distribution shift on real payloads,
-  and validate on your own traffic before relying on it.
-- Keep a deterministic backstop: sanctions hits and invalid signatures should be rejected by code,
-  whatever the model says.
+### 🦙 Quickstart with Ollama
 
-## Training
+Run locally via the official Ollama registry:
 
-LoRA (r=32, alpha=64, all projections, 4-bit NF4 base during training), 1 epoch, lr 1e-4 cosine, effective batch 16,
-max length 3072, loss on the verdict only. Trained with TRL `SFTTrainer`.
+```bash
+ollama run sriram1983007/sra-riskgate