AI Paper Atlas

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

Holdout-Loss-Based Data Selection for LLM Finetuning via In-Context Learning

Ling Zhang, Xianliang Yang, Juwon Yu, Park Cheonyoung, Lei Song, Jiang Bian

ICLR 2026 · Large language models › Instruction and fine-tuning · approach: Efficient LLM fine-tuning systems

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
reweight SFT examples by ICA scoreSFT LLaMA-3-8B · Yahoo_Answers_Topicswin rate vs no reweighting: 85.10%Improved
reweight DPO examples by ICA scoreDPO LLaMA-3-8B · SHP-2win rate vs no reweighting: 77.20%Improved
ICA instead of RHO-Loss (no reference model)reweighting · SFT/DPO/SimPOwin rate vs RHO-Loss: about 46-56%No clear effect
ICA instead of one-shot learning scoringreweighting · all settingswin rate vs One-Shot: 49.60-66.93%Improved
use top-k=1 or 10 holdout demonstrations instead of 3 (ablation)ICA reweighting (LLaMA-3-3B) · Yahoo_Answers_Topicswin rate vs default: 43.50% (k=1), 46.07% (k=10)No clear effect
increase score updates R from 1 to 5 (ablation)ICA reweighting (LLaMA-3-3B) · Yahoo_Answers_Topicswin rate vs default: 52.77%Improved
increase R to 9 (ablation)ICA reweighting (LLaMA-3-3B) · Yahoo_Answers_Topicswin rate vs default: 51.60%No clear effect
hard percentile filtering instead of reweighting (ablation)ICA scores (LLaMA-3-3B) · Yahoo_Answers_Topicswin rate vs reweighting: 40.80% (p50), 48.67% (p75), 40.07% (p90)Hurt
use stronger bge-m3 embeddings for kNN (ablation)ICA reweighting (LLaMA-3-3B) · Yahoo_Answers_Topicswin rate vs default: 52.13%Improved
ICA reweighting on corrupted GSM8K CoTSFT LLaMA-3-8B · GSM8K (40% corrupted)Pass@1: 52.75 -> 79.95Improved
ICA vs RHO-Loss on LLaMA-3-8B GSM8K corruptedreweighting · GSM8K (40% corrupted)Pass@1: 79.50 -> 79.95No clear effect
ICA vs LESS / GREATS selectionSFT LLaMA-3-8B · Yahoo_Answers_Topicswin rate: 52.03% vs GREATS, 51.40% vs LESSNo clear effect

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