# Learning What to Predict: Downstream-Guided Task Design for Continued Pretraining

Source: [Learning What to Predict: Downstream-Guided Task Design for Continued Pretraining](<https://www.sota2.com/research/paper/697c3b7ca8b01e76cc8da800>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-16T05:00:05.515Z

Authors: Shuqi Ke, Giulia Fanti

Published: 2026-01-29T18:38:09.000Z

## Abstract

Continued pretraining is optimized with fixed self-supervised tasks but selected by downstream performance, creating a coarse feedback loop in which practitioners evaluate checkpoints, change data mixtures or objectives, and restart runs, while individual updates remain blind to target capabilities. We ask whether a small set of verifiable downstream examples can provide step-level feedback without directly supervising the learner. We introduce V-pretraining, which decouples a learner trained only with a self-supervised loss from a lightweight task designer that constructs targets or views for unlabeled batches. Given the current learner and batch, V-pretraining scores a candidate construction by predicting the first-order reduction in downstream loss after the induced self-supervised update. The designer maximizes this value; the learner then applies the update with targets or views detached, so downstream labels never update learner parameters. We instantiate V-pretraining as adaptive top-K soft targets for language modeling and learned views or masks for self-supervised vision. Across both modalities, V-pretraining improves target capabilities without degrading generalization. Under wall-clock-matched continued pretraining, it improves GSM8K Pass@1 for Qwen models using 1,024 GSM8K examples only as feedback, including a +7.4 point single-run gain for Qwen2.5-0.5B. In vision, it improves DINOv3 transfer to ADE20K semantic segmentation and NYUv2 depth estimation while preserving ImageNet linear accuracy, suggesting that feedback-guided task construction can improve target capabilities without collapsing general-purpose representations.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2601.22108>)

[PDF](<https://arxiv.org/pdf/2601.22108.pdf>)