Artificial Intelligence & Machine Learning

Claim-Level Evidence Verification for Detecting Unsupported Statements in RAG Responses

A claim-level verification framework to detect unsupported statements in RAG-generated responses.

Claim-Level Evidence Verification for Detecting Unsupported Statements in RAG Responses

Project Information

Category: Artificial Intelligence & Machine Learning

Created: August 2026

Project Description

Researcher: Akshat | Final-Year B.Tech, Computer Science, PSIT, Kanpur Developing a claim-level verification framework to identify unsupported or insufficiently grounded statements generated by retrieval-augmented generation (RAG) systems. Status: Ongoing

Related Projects

A Deep Learning-Based Compression-Robust Framework for Deepfake Detection with Cross-Dataset Generalization and Explainability
A Deep Learning-Based Compression-Robust Framework for Deepfake Detection with Cross-Dataset Generalization and Explainability

A robust deepfake detection framework focused on compression resilience, cross-dataset generalization, and explainability.

View Project
VBRL Assistant