AI Paper Atlas

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

RICo: Refined In-Context Contribution for Automatic Instruction-Tuning Data Selection

Yixin Yang, Qingxiu Dong, Linli Yao, Fangwei Zhu, Weilin Luo, Bin Wang, Zhifang Sui

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Efficient LLM fine-tuning systems · 3 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
train on 15% RICO-selected Alpaca instead of full dataLLaMA3.1-8B · 12-benchmark avgaverage score: 37.95 -> 43.37Improved
train on 15% RICO-selected Alpaca instead of full dataQwen2.5-3B · 12-benchmark avgaverage score: 46.65 -> 47.89Improved
train on RICO-selected Alpaca subsets instead of full dataLLaMA2-7B · 12-benchmark avgaverage score: 38.54 -> 39.19 (1%); 38.45 at 15%No clear effect
select with RICO instead of LESSLLaMA3.1-8B · Avg of benchmarks, 15%average score: 41.89 -> 43.37Improved
select with RICO instead of LESSLLaMA3.1-8B · Pairwise winning scorepairwise winning score: 1.0019No clear effect
increase RICO-selected data beyond 15% (ablation)LLaMA3.1-8B · Alpaca average scoreaverage scoreHurt
apply Alpaca-trained RICO selector to WizardLM-70K (5%)LLaMA3.1-8B · WizardLM avgaverage score: 42.94 -> 47.48Improved
keep 15% RICO subset of WizardLMLLaMA3.1-8B · WizardLM avgaverage score: 42.94 -> 44.18Improved
add continuous refined score instead of binary labels (ablation)RICO · Avg benchmarks (1% Alpaca)average scoreImproved
add fairness adjustment for length bias (ablation)RICO · Avg benchmarks (1% Alpaca)average scoreImproved
RICO selection vs choosing hardest (Top PPL) samplesLLaMA3.1-8B · Avg of benchmarks, 15%average score: 36.80 -> 43.37Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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