UCS: Estimating Unseen Coverage for Improved In-Context Learning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913027236823040 |
|---|---|
| author | Xin, Jiayi Li, Xiang Qiang, Evan He, Weiqing Shang, Tianqi Su, Weijie J. Long, Qi |
| author_facet | Xin, Jiayi Li, Xiang Qiang, Evan He, Weiqing Shang, Tianqi Su, Weijie J. Long, Qi |
| contents | In-context learning (ICL) performance depends critically on which demonstrations are placed in the prompt, yet most existing selectors prioritize heuristic notions of relevance or diversity and provide limited insight into the coverage of a demonstration set. We propose Unseen Coverage Selection (UKS), a training-free, subset-level coverage prior motivated by the principle that a good demonstration set should expose the model to latent cluster unrevealed by the currently selected subset. UCS operationalizes this idea by (1) inducing discrete latent clusters from model-consistent embeddings and (2) estimating the number of unrevealed clusters within a candidate subset via a Smoothed Good--Turing estimator from its empirical frequency spectrum. Unlike previous selection methods, UCS is coverage-based and training-free, and can be seamlessly combined with both query-dependent and query-independent selection baselines via a simple regularized objective. Experiments on multiple intent-classification and reasoning benchmarks with frontier Large Language Models show that augmenting strong baselines with UCS consistently improves ICL accuracy by up to 2-6% under the same selection budget, while also yielding insights into task- and model-level latent cluster distributions. Code is available at https://github.com/Raina-Xin/UCS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_12015 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UCS: Estimating Unseen Coverage for Improved In-Context Learning Xin, Jiayi Li, Xiang Qiang, Evan He, Weiqing Shang, Tianqi Su, Weijie J. Long, Qi Machine Learning Computation and Language In-context learning (ICL) performance depends critically on which demonstrations are placed in the prompt, yet most existing selectors prioritize heuristic notions of relevance or diversity and provide limited insight into the coverage of a demonstration set. We propose Unseen Coverage Selection (UKS), a training-free, subset-level coverage prior motivated by the principle that a good demonstration set should expose the model to latent cluster unrevealed by the currently selected subset. UCS operationalizes this idea by (1) inducing discrete latent clusters from model-consistent embeddings and (2) estimating the number of unrevealed clusters within a candidate subset via a Smoothed Good--Turing estimator from its empirical frequency spectrum. Unlike previous selection methods, UCS is coverage-based and training-free, and can be seamlessly combined with both query-dependent and query-independent selection baselines via a simple regularized objective. Experiments on multiple intent-classification and reasoning benchmarks with frontier Large Language Models show that augmenting strong baselines with UCS consistently improves ICL accuracy by up to 2-6% under the same selection budget, while also yielding insights into task- and model-level latent cluster distributions. Code is available at https://github.com/Raina-Xin/UCS. |
| title | UCS: Estimating Unseen Coverage for Improved In-Context Learning |
| topic | Machine Learning Computation and Language |
| url | https://arxiv.org/abs/2604.12015 |