Structural verification for graph databases. 5M vertices. 35 microseconds. Zero ML.
A semantic network and reasoning engine in which rules, facts and numbers are nodes of one graph, so a statement is itself a node. Importing the statements of the 1.7 TB Wikidata dump builds 983 million nodes on a single machine, and reasoning over them has surfaced thousands of consistency violations in the Wikidata ontology.
Internet never forgots and now thought police never fails
[EMNLP 2025] Think Wider, Detect Sharper: Reinforced Reference Coverage for Document-Level Self-Contradiction Detection.
A tool for auditing bias through large language models
Static analyzer for AI system prompts: parses a prompt into normative rules and reports contradictions, coverage gaps, priority ambiguities, and absolute-rule edge cases - no LLM calls. Deterministic pure-Python lint with CLI, Python API, and CI exit codes. pip install rule-audit
Hybrid memory architecture combining exact recall with infinite-capacity fuzzy understanding for LLMs. Temporal Belief Graph (TBG) for contradiction detection.
GapTime: bi-temporal knowledge-graph memory for production agents. Every fact carries valid time and transaction time, so you time-travel to what was known at any instant, supersede outdated facts without losing history, and detect contradictions. Zero dependencies, TypeScript, edge-ready.
MindGraph OS: An AI powered knowledge operating system that transforms documents and codebases into a living temporal knowledge graph with autonomous reasoning agents.
Independent CLI benchmark harness for agent-memory backends (MemPalace, Mem0, Zep, OpenViking); publishes raw eval logs.
A python script that uses Spacy library to detect contradiction among input statements
MedTrace is a PubMed-powered RAG application that retrieves, embeds, and analyzes biomedical literature, adding a contradiction detection layer to highlight supporting vs. opposing evidence and an LLM-as-a-judge evaluation module to score response relevance, groundedness, and confidence.
discover the unseen, it is an ai-native research discovery platform.
Evaluating retrieval strategy bias in biomedical scientific claim verification using RAG — comparing dense, BM25, hybrid, and query-reformulation retrieval across Support Recall, Contradiction Recall, and Balance Score on SciFact.
A governed cognition substrate: runtime infrastructure for persistent AI cognitive state under explicit governance. Nine primitives. 146 tests. Research prototype.
🔍 Detect contradictions between AI skills — 15 categories, accumulation mode, bilingual IT/EN
Measuring a multi-agent LLM pipeline for scientific contradiction detection - 76% detection, 0/120 false positives, 35% hypothesis error rate. Published negative results with full code and data.
Disagreement-aware scientific RAG: retrieval that keeps conflicting evidence as distinct, source-linked views. Rust pipeline, metrics and benchmark.
Persistent belief graph from document sequences. Extracts entities/claims, resolves duplicates via two-pass entity resolution (alias + cosine similarity), detects contradictions across sources, answers questions with citations. 936 entities from 173 pages using 9 API calls. MCP server for Claude Desktop/Cursor.
Contradiction Detection using LSTMs
To associate your repository with the contradiction-detection topic, visit your repo's landing page and select "manage topics."