NLI-POWERED FACT VERIFICATION
Supports all document formats including PDF, DOCX, HWP, and HWPX. Analyze every claim in AI-generated text against source documents, automatically detecting and correcting hallucinations.
{ "claim": "South Korea's GDP was approximately $1.7 trillion in 2023.",
"verdict": "supported", "confidence": 0.94 }
{ "claim": "Seoul's population is approximately 15 million.",
"verdict": "contradicted", "confidence": 0.97,
"correction": "Actual Seoul population is approximately 9.5 million." }
{ "claim": "Investing in this fund guarantees returns.",
"verdict": "contradicted", "rule": "CG-002",
"correction": "This fund does not guarantee returns and carries the risk of principal loss." }
// corrected_text generated — contradicted claims auto-corrected based on source evidence
96.8%
Detection Rate (500 claims)?
Auto
Auto-Correction?
<2s
3-Layer Max Latency?
7+
Document Formats?
38
Guardrail Rules?
31
Verification Categories?
DOCUMENT SUPPORT
Natively supports all major document formats including Korean Hangul (HWP/HWPX). Upload any source document for automatic analysis and verification.
Native support for Hancom's HWP (OLE binary) and HWPX (ZIP/XML) formats. Accurately extracts tables, text, and formatting.
Precisely extracts text, tables, and layout with PyMuPDF engine.
Parses paragraphs, tables, and styles from Microsoft Word documents.
Directly analyzes plain text and markdown documents.
Strips scripts and styles from web pages, extracting only the body content.
Uploaded documents are automatically processed: text extraction, semantic chunking, E5 vector embedding, and knowledge graph construction — all in one step.
TECHNOLOGY
Our verification pipeline combines semantic retrieval with neural natural language inference for claim-level accuracy.
38 rules · latency <1ms
Compliance rules (CG-001~028), numerical cross-validation, and hallucination pattern matching for instant detection. Handles 73% of total detections within 1ms.
DeBERTa-v3 Cross-Encoder · latency ~50ms
Classifies claim-evidence pairs using cross-encoder NLI model. Leverages structured evidence from Knowledge Graph for improved accuracy.
DeepSeek/Claude · latency ~2s
Re-verifies neutral claims from NLI using LLM with source evidence. Achieves final detection rate of 96.8%.
DataForSEO SERP API · ~3s latency
Collects real-time web search results as additional evidence beyond uploaded documents. Enables fact-checking with web sources alone — no documents required. Provides source URLs and similarity scores per claim.
LLM-powered · Source-grounded
Automatically generates accurate corrections for contradicted claims based on source document evidence. Preserves the original writing style while fixing only the facts, delivering ready-to-use corrected_text.
Detection Mechanism Contribution
Detection Rate Evolution
5-round improvement on 500-claim benchmark
TruthAnchor v3.2.0 · 500 claims verified · 31 categories
Category Performance
TRY IT
Paste any LLM-generated text to see how HalluGuard's 3-layer verification works.
* This is a simulation demo. Sign in for real verification.
HOW IT WORKS
Upload source documents as verification references. Supports PDF, DOCX, HWP, HWPX, TXT, MD, and HTML. Documents are automatically chunked and vector-embedded.
Automatically extract individual factual claims from LLM output.
Semantic search for relevant evidence using E5-large embedding model.
Classify claims as supported, contradicted, or neutral using DeBERTa-v3 cross-encoder.
Search the web in real-time to collect external sources, combining them with uploaded documents as hybrid evidence.
Automatically correct contradicted claims based on source evidence, generating ready-to-use corrected text.
PRICING
Start free. No credit card required.
Service period: 1 month from the date of payment
Recommended
Pro
Business
Enterprise
Free plan includes 15 verifications per month. No credit card required.
Get Started Free