Atlas: Few-shot Learning with Retrieval Augmented Language Models
by Meta AI / University College London · unknown · Last verified 2026-03-17
Atlas is a retrieval-augmented language model designed for few-shot learning. It uniquely pre-trains its retriever and language model components jointly, enabling it to effectively leverage external knowledge documents. This approach allows Atlas to achieve state-of-the-art few-shot performance on knowledge-intensive NLP benchmarks like MMLU, outperforming much larger models.
https://arxiv.org/abs/2208.03299 ↗B
B—Above Average
Adoption: B+Quality: AFreshness: BCitations: AEngagement: F
Specifications
- License
- CC BY-NC 4.0
- Pricing
- unknown
- Capabilities
- few-shot learning, retrieval-augmented generation (RAG), joint training of retriever and language model, knowledge-intensive NLP tasks, in-context learning with retrieved documents, dense passage retrieval, knowledge grounding for language models, fact verification, open-domain question answering
- Integrations
- Use Cases
- [object Object], [object Object], [object Object], [object Object]
- API Available
- No
- Tags
- rag, few-shot-learning, retrieval-augmented-generation, language-model, joint-training, knowledge-intensive-tasks, dense-retrieval, mmlu-benchmark, natural-language-processing, information-retrieval
- Added
- 2026-03-17
- Completeness
- 1%
Index Score
65.2Adoption
70
Quality
86
Freshness
60
Citations
80
Engagement
0