Almanac
technique

instruction-based multitask pretraining

techniqueactiveinstruction-based-multitask-pretraining-39bd1e66·1 events·first seen 1mo ago

Aliases: instruction-based multitask pretraining

Co-occurring entities

More like this (12)

Recent events (1)

4Qwen Research·1mo ago·source ↗

OFA: Towards Building a One-For-All Unified Multimodal Pretrained Model

Alibaba's Qwen team introduces OFA (One-For-All), a unified multimodal pretrained model designed to handle both understanding and generation tasks across multiple modalities within a single framework. The model is pretrained using instruction-based multitask pretraining to endow it with diverse capabilities. This work was published in late 2022 as part of the broader wave of generalist multimodal models. It represents an early effort toward a single model architecture capable of spanning vision, language, and cross-modal tasks.