LittleLearner¶
Product or Programme · All products and programmes
Evidence · 5 claims (5 for) · 1 study (1 causal),q2· 0 of 1 report an effect size · 5 claims rest on one study
Description¶
LittleLearner is a 5B-parameter language model trained from scratch by Fanfei Li et al. on the K–5 LittleCurriculum corpus for studying knowledge acquisition, transfer, calibration, and extension in language models.
Components¶
Claims¶
- GRPO post-training boosts in-scope K-5 capabilities but fails to recover Beyond-K-5 capabilities, even when post-trained on out-of-scope data [+W]
- LittleLearner's capabilities tightly track its controlled exposure: strong on in-scope tasks, degrading beyond the K-5 boundary [+W]
- Scaling model size improves performance within K-5 and at the boundary but does not recover Beyond-K-5 performance [+W]
- In-context learning with few-shot examples and explanations does not unlock Beyond-K-5 reasoning for the 5B LittleLearner [+W]
- Across scaling, post-training, and in-context learning, the pretraining filter rather than the intervention sets LittleLearner's capability ceiling [+W]
Related Products and Programmes¶
Key Sources¶
- Fanfei Li, Jana Zeller, Manuel Prada-Corral, Thaddäus Wiedemer, Prasanna Mayilvahanan, Ryan Cotterell, Wieland Brendel. (2026). LittleLearner: Language Models Under Pedagogically Controlled Knowledge Exposure. Preprint. https://arxiv.org/abs/2608.13545
Page metadata
| Field | Value |
|---|---|
type |
product |
id |
littlelearner |
title |
LittleLearner |
description |
LittleLearner is a 5B-parameter language model trained from scratch by Fanfei Li et al. |
status |
review |
product_kind |
software |
generated |
by process:settle-candidates on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.