AI๊ฐ ์์ง์ด๊ณ ๋ฏธ๋๊ฐ ์
๋ฐ์ดํธ๋ฉ๋๋ค.DenseOn
> ๐ฏ **TL;DR**: A 149M-parameter dense (single-vector) retrieval model achieving **56.20 nDCG@10 on BEIR**, the first sub-150M dense model to break the 56 mark, **topping all base-size dense models** and outperforming several models **4ร larger**. State-of-the-art retrieval is increasingly dominated by closed models, either hidden behind APIs or trained on undisclosed data. This blocks reproducibility, prevents study of possible data leakage, and gatekeeps progress to a handful of private labs. We thus decided to gather and curate a large amount of data and explore various mixtures. We release all the data used in our explorations: - Gathered pre-training data, 1.4B query-documents pairs alongside annotations used for non-destructive filtering (structural filtering, deduplication, cross-encoder pair relevancy) - Best pre-training mixture found with already applied filters - Fine-tuning datasets with query, positive and 2048 mined documents alongside their scores for 1.88M samples. Based on our findings, we trained LateOn (multi-vector/ColBERT) and DenseOn (single vector/dense) models on a proprietary Apache 2.0-compatible training dataset and release those models as well. Both are built on the ModernBERT backbone at 149M parameters, a size we believe sits at the sweet spot: large enough to handle real-world queries and documents, small enough to serve at high throughput in latency-sensitive production systems. For more information, please read our blogpost.
๋ฒ์ญ ์ค๋น ์ค์ ๋๋ค. ์๋ฌธ์ ํ์ํฉ๋๋ค.
> ๐ฏ **TL;DR**: A 149M-parameter dense (single-vector) retrieval model achieving **56.20 nDCG@10 on BEIR**, the first sub-150M dense model to break the 56 mark, **topping all base-size dense models** and outperforming several models **4ร larger**. State-of-the-art retrieval is increasingly dominated by closed models, either hidden behind APIs or trained on undisclosed data. This blocks reproducibility, prevents study of possible data leakage, and gatekeeps progress to a handful of private labs. We thus decided to gather and curate a large amount of data and explore various mixtures. We release all the data used in our explorations: - Gathered pre-training data, 1.4B query-documents pairs alongside annotations used for non-destructive filtering (structural filtering, deduplication, cross-encoder pair relevancy) - Best pre-training mixture found with already applied filters - Fine-tuning datasets with query, positive and 2048 mined documents alongside their scores for 1.88M samples. Based on our findings, we trained LateOn (multi-vector/ColBERT) and DenseOn (single vector/dense) models on a proprietary Apache 2.0-compatible training dataset and release those models as well. Both are built on the ModernBERT backbone at 149M parameters, a size we believe sits at the sweet spot: large enough to handle real-world queries and documents, small enough to serve at high throughput in latency-sensitive production systems. For more information, please read our blogpost.