AI Paper Atlas

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

Not All Documents Are What You Need for Extracting Instruction Tuning Data

Chi Zhang, Huaping Zhong, Hongtao Li, Chengliang Chai, Hongjiawei, Yu-Ping Wang, Yuhao Deng, Jiacheng Wang, Yizhou Yan, Qiu Jiantao, Conghui He, Lei Cao

ICLR 2026 · Large language models › Instruction and fine-tuning · approach: Instruction data selection · 2 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
add contrastive-learning warm-up for document embeddings (ablation)EQUAL · GSM8K (Full FT, 5% docs)accuracy: 69.73 -> 73.01Improved
add multi-armed bandit iterative selection (ablation)EQUAL · GSM8K (Full FT, 5% docs)accuracy: 71.90 -> 73.01Improved
use optimal transport reward instead of average similarity (ablation)EQUAL · GSM8K (Full FT, 5% docs)accuracy: 70.77 -> 73.01Improved
add contrastive-learning warm-up (ablation)EQUAL · HumanEval (Full FT, 5% docs)pass rate: 44.3 -> 49.4Improved
add multi-armed bandit iterative selection (ablation)EQUAL · HumanEval (Full FT, 5% docs)pass rate: 46.9 -> 49.4Improved
use optimal transport reward (ablation)EQUAL · HumanEval (Full FT, 5% docs)pass rate: 47.6 -> 49.4Improved
add contrastive-learning warm-up (ablation)EQUAL · GSM8K (LoRA, 5% docs)accuracy: 64.05 -> 67.32Improved
use 1000 clusters instead of k=50 (ablation)EQUAL · GSM8K and MATHaccuracy: 2.90% and 2.05% lower accuracy with k=50Improved
use 1000 clusters instead of k=50,000 (ablation)EQUAL · GSM8K and MATHaccuracy: 2.56% and 1.79% lower accuracy with k=50,000Improved
swap k-means for BIRCH or DBSCAN clustering (ablation)EQUAL · downstream tasksaccuracyNo clear effect
raise extracted document ratio from 5% to 20%EQUAL · MATH (Full FT)accuracy: 35.10 -> 41.40Improved
use EQUAL instead of Influence selectionLlama-3-8B Full FT · GSM8K (Full FT, 5% docs)accuracy: 65.20 -> 73.01Improved

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