nodd-repo/comment-moderation

comment_moderation (v4)

Decide whether a user comment on a product blog is acceptable to publish.

A small classifier trained with nodd, exported as ONNX fp32 and int8 for browser and Node.js inference. This version uses a larger offline synthetic training dataset.

Label	Meaning
ok	Normal comment. Can be positive, negative or off topic, but harmless.
spam	Ads, links to unrelated products or services, SEO junk, scams.
toxic	Insults, harassment, threats or hate toward people or groups.

Evaluation

Metric	Value
Test macro F1, PyTorch	0.9339
Test macro F1, int8 ONNX	0.9345
int8 vs PyTorch label agreement	97.28%
int8 download	24.31 MB
Train / validation / test	4197 / 257 / 257
Temperature	1.299944
Confidence threshold	0.730609
Validation precision at threshold	97.15%
Test accuracy on covered inputs, PyTorch	95.56%
Test coverage, PyTorch	96.50%
Training seed	42

See baseline comparison, evaluation report, and browser parity results.

Data and limitations

Added 1,000 examples per label using deterministic offline templates, with generation seed 20260928. Labels were assigned by construction, not by an independent teacher. Template variants are correlated and do not represent independent scenarios. New rows are training-only; the original validation and test records were preserved. Exact duplicates and examples with embedding cosine similarity above 0.90 to either holdout were rejected.

All evaluation inputs are synthetic; real-world accuracy is unmeasured. More data did not consistently improve held-out quality: consult the comparison before replacing an earlier version. The 97% precision target selected on validation was not met on this version's covered PyTorch test subset. Quantization can change individual probabilities substantially, so label parity does not establish confidence equivalence. Prompt-injection detection, where applicable, is not a standalone security boundary.

Use

Download this version and serve it from your own origin:

hf download nodd-repo/comment-moderation --revision v4 --local-dir public/models/comment_moderation/v4
npm install @nodd/browser

import { nodd } from "@nodd/browser";
const model = await nodd.load("/models/comment_moderation/v4");
const decision = await model.decide("Your input text");
if (!model.isConfident(decision)) { /* escalate for review */ }
model.dispose();

For Node.js, use @nodd/node and the downloaded directory path. Requires nodd JavaScript packages >= 0.2.0.

Artifacts

• onnx/model_quantized.onnx, onnx/model.onnx: int8 and fp32 browser models.
• nodd.json, tokenizer files and config.json: browser runtime configuration.
• encoder/: original Transformers checkpoint, loadable with AutoModelForSequenceClassification.from_pretrained(local_path, subfolder="encoder").
• labeled.jsonl: exact training, validation and test records with split and provenance fields.
• synthetic_manifest.json: generation metadata and baseline fingerprint.
• model_card.json, reports and parity files: training, calibration and evaluation evidence.

Earlier releases remain available through their version tags. The generation script and cross-task results are in the source repository.