Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞

Local, privacy-preserving semantic document search engine with citation verification

Research Agora: Claude Code skills, benchmarks & tools for ML researchers — paper writing, citation verification, experiment tracking, LaTeX automation

Automated citation verification system with Agentic self-learning. Validates academic metadata and semantic consistency using ML and multi-source databases. 具備 Agentic 自學能力的自動化學術引用驗證系統。整合多方資料庫與 ML 技術，精準確保文獻元資料及語義內容的一致性。

Enhance research paper writing by organizing workflows, improving structure, and supporting self-review for clear, logical, and well-aligned content.

HALLMARK: Citation hallucination detection benchmark for ML papers — 2,525 entries, 14 hallucination types, 3 difficulty tiers, 10 baselines including LLMs and verification tools

Open-source AI reasoning auditor for legal citations. Verifies existence, quote accuracy, and logical coherence against primary sources.

AI-powered citation verification tool that catches fake papers, fixes broken references, and validates academic sources against 200M+ papers. Built with Next.js + TypeScript.

AI agent that verifies whether citations in documents truly support their claims. Parses docs, extracts claims, fetches sources, and assigns verdicts (SUPPORTED, NOT_SUPPORTED, etc.) with explanations. CLI, Python API, and REST API available.

本项目主要实现文献分类以及自动生成可溯源相关工作

A LangGraph-based reasoning engine with enforced citations for RAG.

Automated research quality assurance. Catches fabricated citations, overclaimed results, irreproducible numbers.

Audit-Ready RAG with Citation Enforcement and Reliability Evaluation — COMP3931 Dissertation, University of Leeds

A dataset that includes citations (source text and referenced text)

Automate Claude Code optimization using constraints, metrics, and autonomous iterations to drive continuous improvement in research workflows.

Build autonomous ML research in Elixir: design, train, and iterate GPT models across GPUs with fault-tolerant BEAM concurrency

Port Apple Silicon support for Karpathy’s autoresearch to run fixed-time training loops natively without PyTorch or CUDA dependencies.

Forensic reference-integrity auditor for academic publishing. Prompt-engineered deep-scan verification using Claude with live web search.

Organize genealogy research with structured AI prompts, vault templates, and workflows for source-backed family history work

Automate code improvement by detecting issues, fixing bugs, and simplifying code on separate branches before merging to main.