AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

From Macro to Micro: Probing Dataset Diversity in Language Model Fine-Tuning

Haoyu Li, Xuhong Li, Yiming Dong, Kun Liu

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Efficient LLM fine-tuning systems · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase token-level (micro) diversity of responsesSFT data, Llama-2-7B · AlpacaEval 2.0 + Arena Hard (10K/20K/30K)score slope: slope 5.35 / 6.68 / 10.06 (x10^-2)Improved
increase token-level (micro) diversity of instructionsSFT data, Llama-2-7B · AlpacaEval 2.0 + Arena Hard (10K/20K/30K)score slope: slope -5.45 / -2.83 / -1.80 (x10^-2)Hurt
increase tag-level (meso) diversitySFT data, Llama-2-7B · AlpacaEval 2.0 + Arena Hardscore slope: slope 4.93 vs macro 2.37 on instructions (10K)Improved
increase semantic (macro) diversity of instructionsSFT data, Llama-2-7B · AlpacaEval 2.0 + Arena Hardscore slope: slope 2.37 (10K)Improved
increase dataset size from 10K to 30KSFT data, Llama-2-7B · AlpacaEval 2.0 + Arena HardscoreImproved
apply micro response diversity to Llama-2-70B and Llama-3-8B (ablation)micro response strategy · 10K, AlpacaEval 2.0 + Arena HardscoreImproved
control response length to about 500 tokens (ablation)micro response strategy · 10K, AlpacaEval 2.0 + Arena Hardscore vs diversity relationNo clear effect
use word segmentation instead of model tokenizer for micro diversity (ablation)micro response strategy · 10K, AlpacaEval 2.0 + Arena HardscoreHurt
use information entropy of response tokens as diversity metricdiversity metrics · 10K datasetsPearson correlation with score: 0.58Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy