data-selection 开源项目推荐（78 个）

OpenDCAI /DataFlex

DataFlex：面向大语言模型以数据为中心训练的统一基准与评估平台

OpenDataBox /awesome-data-llm

Official Repository of "LLM × DATA" Survey Paper

princeton-nlp /LESS

[ICML 2024] LESS: Selecting Influential Data for Targeted Instruction Tuning

p-lambda /dsir

DSIR large-scale data selection framework for language model training

alon-albalak /data-selection-survey

A Survey on Data Selection for Language Models

HKUSTDial /LEAD

🔥[VLDB'26] Official repository for the paper "LEAD: Iterative Data Selection for Efficient LLM Instruction Tuning".

georgian-io /Transformers-Domain-Adaptation

⛔ [DEPRECATED] Adapt Transformer-based language models to new text domains

gszfwsb /Data-Whisperer

Code for ACL 2025 Main paper "Data Whisperer: Efficient Data Selection for Task-Specific LLM Fine-Tuning via Few-Shot In-Context Learning".

waltonfuture /InstructionGPT-4

InstructionGPT-4

zhuang-li /SCAR

[ACL 2025 main] SCAR: Data Selection via Style Consistency-Aware Response Ranking for Efficient Instruction-Tuning of Large Language Models

yichengchen24 /ACP

[CVPR25] Official repository for Auto Cherry-Picker: Learning from High-quality Generative Data Driven by Language

princeton-pli /STAT

[ICLR 2026] Skill-Targeted Adaptive Training

yichengchen24 /MIG

[ACL25 Findings] Official repository for MIG: Automatic Data Selection for Instruction Tuning by Maximizing Information Gain in Semantic Space

USTC-StarTeam /ZIP

arXiv 2024 | ZIP: entropy-law data selection for efficient LLM alignment.

WisdomShell /ADG

[ACL'26 Main Conference] Instruction Data Selection via Answer Divergence

yueyu1030 /Patron

该项目在 GitHub 上没有填写描述

ZifanL /TSDS

Implementation of TSDS: Data Selection for Task-Specific Model Finetuning. An optimal-transport framework for selecting domain-specific and task-specific training data to improve LLM finetuning and instruction tuning.

StigLidu /GradAlign

[COLM 2026] Official implementation for "GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning"

reds-lab /projektor

This is an official repository for "Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources" (NeurIPS 2023).

Nokia-Bell-Labs /data-centric-federated-learning

Enhancing Efficiency in Multidevice Federated Learning through Data Selection