AI Paper Atlas

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

A Critical Look at Targeted Instruction Selection: Disentangling What Matters (and What Doesn’t)

Nihal V. Nayak, Paula Rodriguez-Diaz, Neha Hulkund, Sara Beery, David Alvarez-Melis

ICML 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning generalization and delta compression · 0 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
use LESS gradient-based representation instead of RDS+ or EMBED for greedy round-robin selection (ablation)Llama-2-7B, Tulu V2 pool, round-robin selection · BBH, Codex, GSM8K, TyDiQA, MMLU-Pro query lossquery-set cross-entropy lossImproved
use LESS representation: subset-query distance quantile vs performance (ablation)Llama-2-7B, round-robin selection · five target tasks, 10 distance quantilesquery loss and downstream performance correlation with distanceImproved
use RDS+ or EMBED representations (distance to query as predictor) (ablation)Llama-2-7B, round-robin selection · five target tasks, 10 distance quantilesSpearman correlation of distance with query lossNo clear effect
lower query loss from LESS representation (ablation)LESS (RR) vs EMBED (RR) · first distance quantile, five target tasksdownstream performanceMixed
LESS (RR) vs other representations across budgets (ablation)Llama-2-7B, Tulu V2 pool · downstream tasks, budgets 500 to 10000task accuracyMixed
targeted selection (RDS+/EMBED) instead of random sampling (ablation)Llama-2-7B · larger budgets, TyDiQA and MMLU-Prodownstream performanceNo clear effect
greedy round-robin selection at small budgets (ablation)LESS, Llama-2-7B · low budgets, five target tasksdownstream performanceImproved
use UOT or KNN-KDE instead of round-robin at large budgets (ablation)LESS, Llama-2-7B · high budgets, five target tasksdownstream performanceImproved
use doubly greedy (DG) selection instead of other algorithms (ablation)LESS, Llama-2-7B · three of five datasetsdownstream performance and query lossHurt
use UOT selection on LESS representations (ablation)LESS, Llama-2-7B · five target tasksquery lossImproved
switch candidate pool from Tulu V2 to Dolci Instruct (ablation)targeted selection methods · downstream tasksdownstream performanceMixed
increase selection budget (ablation)LESS variants vs random · MMLU-Proloss gap to random samplingHurt

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