THUDM/LongCite
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA
Implements a coarse-to-fine (CoF) pipeline for automatically constructing sentence-level citation annotations in long-context QA datasets, enabling models to generate verifiable responses with precise source references. Provides fine-tuned GLM-4-9B and Llama-3.1-8B variants supporting 128K context windows, along with the LongBench-Cite benchmark for evaluating citation quality and response correctness using GPT-4o as an automated judge.
519 stars. No commits in the last 6 months.
Stars
519
Forks
31
Language
Python
License
Apache-2.0
Category
Last pushed
Dec 31, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/THUDM/LongCite"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
MadryLab/context-cite
Attribute (or cite) statements generated by LLMs back to in-context information.
microsoft/augmented-interpretable-models
Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.
Trustworthy-ML-Lab/CB-LLMs
[ICLR 25] A novel framework for building intrinsically interpretable LLMs with...
poloclub/LLM-Attributor
LLM Attributor: Attribute LLM's Generated Text to Training Data
UKPLab/5pils
Code associated with the EMNLP 2024 Main paper: "Image, tell me your story!" Predicting the...