Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · TrustSafeAI

RADAR-Vicuna-7B

RADAR-Vicuna-7B is a RoBERTa-based AI-text detector from TrustSafeAI. According to the model card, it was trained with adversarial learning between a detector and a paraphraser.

Publisher
TrustSafeAI
Task
text-classification
Model type
roberta
License
Unknown
Library
transformers
Publication status
Accepted · not indexed

Model overview

RADAR-Vicuna-7B is published by TrustSafeAI as a text-classification model. The captured configuration identifies RobertaForSequenceClassification with model type roberta. According to the model card, it detects AI-generated text after adversarial training on OpenWebText human text and Vicuna-7B-generated AI text.

Recorded capabilities

Adversarial detector training

According to the model card, a paraphraser tries to make AI text more human-like while the detector improves its ability to identify AI text.

RoBERTa encoder classifier

The model card describes an encoder-only transformer based on RoBERTa, matching the captured RobertaForSequenceClassification architecture.

Vicuna-7B-derived AI data

According to the model card, AI-text training data was generated by Vicuna-7B completing OpenWebText prefix spans.

Documented non-commercial terms

According to the model card, the model carries a non-commercial license inherited from Vicuna-7B-v1.1 and is not for commercial activities.

Use cases in the source record

  • Assisted AI-text screening workflows that classify text as human-written or model-generated, with further validation before evidentiary use.

Limitations and unknowns

  • No parameter count was extracted for this record.
  • No evaluation results were extracted from this record.
  • According to the model card, RADAR cannot always make correct predictions, so detection results need further validation before evidentiary use.
  • License terms are publisher-described as non-commercial and have not been independently verified by Ethen.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: TrustSafeAI/RADAR-Vicuna-7B

Captured: Unknown. Processed: 2026-09-07T19:34:38.000882+00:00.

RADAR Model Card Model Details RADAR-Vicuna-7B is an AI-text detector trained via adversarial learning between the detector and a paraphraser on human-text corpus ( OpenWebText ) and AI-text corpus generated based on OpenWebText . Developed by: TrustSafeAI Model type: An encoder-only language model based on the transformer architecture (RoBERTa). License: Non-commercial license (inherited from Vicuna-7B-v1.1) Trained from model: RoBERTa Model Sources Project Page: https://radar.vizhub.ai/ Paper: https://arxiv.org/abs/2307.03838 IBM Blog Post: https://research.ibm.com/blog/AI-forensics-attribution Uses Users could use this detector t…

F001F002F003F004F005F006F007F008F010F011F012F013F014F015F016F017