Skip to main content

Research Citations

Every model, method, and dataset underpinning Aiscern is grounded in peer-reviewed research. We cite all work we build on — transparency is a core part of how we operate.

Text Detection

Binoculars: Zero-Shot Detection of LLM-Generated Text

Hans et al. · ICML 2024

Cross-perplexity scoring between two observer LLMs — the core of our text detection signal.

RoBERTa: A Robustly Optimized BERT Pretraining Approach

Liu et al. · arXiv 2019

Backbone for the openai-detector fine-tune used in our text ensemble.

Detecting LLM-Generated Text in the Wild with Zero-Shot Methods

Koike et al. · ACL 2024

Evaluation of perplexity-based zero-shot detectors — informed our threshold calibration.

M4: Multi-generator, Multi-domain, Multi-lingual Black-Box Machine-Generated Text Detection

Wang et al. · EACL 2024

Benchmark dataset and methodology for cross-generator detection — used in our text benchmarks.

RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors

Dugan et al. · ACL 2024

Adversarial robustness evaluation framework — used to stress-test our text models.

Image Detection

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (ViT)

Dosovitskiy et al. · ICLR 2021

Foundation architecture for our image detection classifier.

Learning Transferable Visual Models From Natural Language Supervision (CLIP)

Radford et al. (OpenAI) · ICML 2021

Embedding model for our image RAG pipeline and similarity-based retrieval.

Towards Universal Fake Image Detection Exploiting Style Latent Space

Ojha et al. · CVPR 2023

Zero-shot generalization approach for detecting unseen generators — informs our ensemble strategy.

On the Detection of Synthetic Images Generated by Diffusion Models

Corvi et al. · ICASSP 2023

Frequency-domain forensics for diffusion model outputs — basis for our spectral signal layer.

Audio Detection

wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

Baevski et al. · NeurIPS 2020

Foundation model for our audio deepfake detection fine-tune.

ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild

Yamagishi et al. · ASVspoof 2021

Primary evaluation dataset and challenge protocol for our audio models.

ADD 2023: The Second Audio Deepfake Detection Challenge

Yi et al. · ICASSP 2023

Extended benchmark including emotional and singing deepfakes.

Robust Audio Anti-Spoofing with Fusion-Reconstruction Learning on Multi-Order Spectrograms

Chen et al. · Interspeech 2022

Spectrogram fusion approach informing our multi-feature audio pipeline.

Video / Deepfake Detection

FaceForensics++: Learning to Detect Manipulated Facial Images

Rössler et al. · ICCV 2019

Primary training and evaluation dataset for face manipulation detection.

Towards Generalizable Detection of Face Forgery via Vision Transformer

Zheng et al. · arXiv 2021

ViT-based frame-level deepfake detector used in our video ensemble.

Detecting Deep-Fake Videos from Appearance and Behavior

Agarwal et al. · IEEE Workshops 2020

Temporal behavioral signals for video deepfake detection.

RAG & Ensemble Methods

Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

Lewis et al. · NeurIPS 2020

Foundational RAG paper — architecture behind our vector-retrieval confidence augmentation.

Model Cards for Model Reporting

Mitchell et al. · FAccT 2019

Guidelines we follow for documenting bias, limitations, and intended use of all models.

pgvector: Open-Source Vector Similarity Search for Postgres

pgvector contributors · GitHub

Vector database extension used for pattern storage and cosine similarity retrieval.

Missing a citation?

If you believe we are using your work without attribution, please contact research@aiscern.com and we will add it promptly.

Also see: Benchmarks · Methodology