hitz-zentroa/GoLLIE
GoLLIE is a transformer-based LLM fine-tuned to follow annotation guidelines for zero-shot named-entity recognition, relation extraction, and event extraction.
Not currently ranked — collecting fresh signals.
star history
GoLLIE is a guideline-following Large Language Model designed for information extraction tasks including named-entity recognition, relation extraction, and event extraction. Unlike typical zero-shot approaches that rely on pre-encoded knowledge, GoLLIE is trained to interpret and follow detailed annotation schemas defined on the fly. The model can perform inference with user-defined schemas without prior training, making it adaptable for varied extraction requirements.
Frequently asked
- What is hitz-zentroa/GoLLIE?
- GoLLIE is a transformer-based LLM fine-tuned to follow annotation guidelines for zero-shot named-entity recognition, relation extraction, and event extraction.
- Is GoLLIE open source?
- Yes — hitz-zentroa/GoLLIE is open source, released under the Apache-2.0 license.
- What language is GoLLIE written in?
- hitz-zentroa/GoLLIE is primarily written in Python.
- How popular is GoLLIE?
- hitz-zentroa/GoLLIE has 441 stars on GitHub.
- Where can I find GoLLIE?
- hitz-zentroa/GoLLIE is on GitHub at https://github.com/hitz-zentroa/GoLLIE.