AI Paper Atlas

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

SMILE: Extended Deep Submodular Function-Based Instruction and In-context Learning Demonstration Selection

Zihan Chen, Chengshuai Shi, Song Wang, Jundong Li, Cong Shen

ICML 2026 · Large language models › Instruction and fine-tuning · approach: Instruction data selection

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 query-relevance expert f_SQ (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 58.5 -> 63.0Improved
add query-instruction scorer s_QI (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 58.0 -> 63.0Improved
add diversity expert f_SS (ablation)SMILE (Qwen3-4B-Instruct) · Dateaccuracy: 44.8 -> 63.2Improved
add diversity expert f_SS (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 59.1 -> 63.0Improved
add instruction-compatibility expert f_SI (ablation)SMILE (Qwen3-4B-Instruct) · XSumROUGE-L F1: 17.6 -> 18.5Improved
add instruction-compatibility expert f_SI (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 61.8 -> 63.0Improved
use hard min aggregation instead of mean (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 58.5 -> 63.0Improved
use hard min at inference instead of softmin (ablation)SMILE (Qwen3-4B-Instruct) · six-dataset averageaverage score: 61.0 -> 63.0Improved
joint instruction and demonstration selection (SMILE) instead of APE+R.S.prompt optimization (Qwen3-4B-Instruct) · six-dataset averageaverage score: 59.8 -> 63.0Improved
joint instruction and demonstration selection (SMILE) instead of APE+R.S.prompt optimization (Llama3.1-8B-Instruct) · six-dataset averageaverage score: 57.4 -> 60.7Improved
transfer surrogate trained on Qwen feedback to Gemini2.5-FlashSMILE-Q · five-dataset averageaverage score: 78.8 -> 83.7Improved
SMILE-L vs strongest baseline APE+R.S. on Gemini2.5-FlashSMILE-L · five-dataset averageaverage score: 83.1 -> 83.8Improved

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