Spaces:
Running on Zero
Running on Zero
|
Download README.md from VanillaVector/Hybrid-Comment-Moderation: direct link, hf CLI and curl.
- Browser
- Download file 1.5 kB
-
https://huggingface.co/spaces/VanillaVector/Hybrid-Comment-Moderation/resolve/main/README.md
- Command line
-
hf download hf://spaces/VanillaVector/Hybrid-Comment-Moderation/README.md
-
curl -L -o README.md https://huggingface.co/spaces/VanillaVector/Hybrid-Comment-Moderation/resolve/main/README.md
1.5 kB
A newer version of the Gradio SDK is available: 6.32.0
metadata
title: Hybrid Comment Moderation
emoji: π¬
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.34.0
app_file: app.py
pinned: false
license: mit
tags:
- content-moderation
- hybrid-ai
- distilbert
- qwen
- cs541
π¬ Hybrid Comment Moderation
CS541 AI Final Project β Stevens Institute of Technology
A two-stage content moderation pipeline that routes comments intelligently:
- Stage 1 β ML Classifier (DistilBERT, CPU, ~7 ms): handles confident predictions instantly.
- Stage 2 β LLM Reviewer (Qwen2.5-14B-Instruct, NF4, GPU, ~1β2 s): reviews only ambiguous comments.
Tabs
| Tab | Description |
|---|---|
| π Live Demo | Type a comment, pick a strategy, see the verdict |
| π Evaluation | Pre-computed metrics on 100 balanced Civil Comments samples |
| ποΈ Architecture | Pipeline diagram and design rationale |
Usage
- ML-only works for everyone (CPU, no sign-in needed).
- Hybrid and LLM-only require a free HuggingFace account for ZeroGPU access.
- First LLM call loads the 14B model (~30β40 s); subsequent calls take ~1β2 s.
Reference
- Textbook: Understanding Deep Learning by Simon J.D. Prince
- Dataset: google/civil_comments
- ML model: martin-ha/toxic-comment-model
- LLM: Qwen/Qwen2.5-14B-Instruct