[ICLR2025 Oral] ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding
Unified schema definition, validation, and JSON generation for Elixir
A TypeScript library for AI-powered JSON mock data generation using schema-based configuration
MockingJar is a modern web app for visual schema building, data generation, and AI-powered recovery. Built with Next.js 15, Material UI, Claude, and the MockingJar Library, it's designed for developers, QA engineers, and data analysts.
JSON generator is a program that creates structured data in JSON (JavaScript Object Notation) format.
QLoRA fine-tuning project for structured cybersecurity threat analysis using small language models.
Sovereign structured LLM generation engine. SGLang RadixAttention + Pydantic schema enforcement + Ollama fallback. Guaranteed JSON output. No hallucinated keys.
Structured JSON output from 12 LLM providers with a single function call. Type-safe, retry-resilient, zero boilerplate.
Constrained decoding for reliable function calling with small language models.
Multimodal VLM fine-tuning (Qwen2.5-VL-3B via LoRA) on FoodExtract-1k-Vision for image-to-JSON attribute parsing and evaluation.
AI-powered cloud cost optimization tool that converts plain-English project descriptions into structured cloud billing analysis and multi-cloud cost optimization recommendations using LLMs.
BibCreator is a Java application designed for processing BibTeX files, which are commonly used for organizing bibliographic references. This tool can take BibTeX files as input and create corresponding JSON files in different formats (IEEE, ACM, NJ) for further analysis and use. It's particularly useful for managing academic references.
A Visual Studio Code tool for easy HTML text extraction and management.
A JSON unit for older Delphi Version i.e. Delphi 7
Production-grade multi-agent content generation system that transforms structured product data into machine-readable pages using modular agents, reusable logic blocks, template engines, and orchestrated automation flows
A comparative evaluation of fine-tuned small language models (LLaMa 3.2, Gemma 4) and a large language model (Qwen2.5-72B) for converting Turkish smart-home commands into Home Assistant JSON service calls using prompt engineering.
A constrained-decoding pipeline that turns prompts into typed function calls.
To associate your repository with the json-generation topic, visit your repo's landing page and select "manage topics."