← 研踪 PaperTrailarXiv:2611.18179 · v1
Explanations that Transfer: Interpreting Medical Retrieval with Sparse AutoEncoders
首次提交 2026-11-13 21:48
版本更新 2026-11-13 21:48
雷达观测 2026-11-16 11:41
摘要
Medical AI systems that answer users' queries rely on retrieving medical knowledge from large document collections. In this context, finetuning general models on medical datasets can improve retrieval relevance. Yet the dense embeddings underlying modern retrieval systems remain largely opaque, making it difficult to understand the mechanics behind their results. Moreover, adapting to a specific medical dataset risks losing transferability to new clinical corpora. In this work, we tackle these issues by applying Sparse Autoencoders (SAEs) to medical information retrieval, transforming dense text embeddings into semantically interpretable representations. To do so, our model combines a reconstruction objective with an InfoNCE contrastive loss that directly preserves query-document relevance. We evaluate three embedding models across six medical retrieval datasets, both in-domain and out-of-domain. Trained on a medical corpus, our sparse codes improve in-domain retrieval over the original embeddings, and lose less out-of-domain performance than their dense counterparts, especially when the target corpus is distant from the training one. Importantly, our experiments show that the method yields interpretable features whose semantic content can be inferred from the queries and documents that activate them most strongly. The interpretation of these features sheds light on why a particular document was retrieved, and transfers beyond the training distribution.
作者备注:Proceedings of Machine Learning Research 346, 2026 (Machine Learning for Health, ML4H 2026). Code: https://github.com/doctolib-lab/medsae
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v3 · 2026-11-18 19:46 · 最新观测版本
Explanations that Transfer: Interpreting Medical Retrieval with Sparse AutoEncoders
Medical AI systems that answer users' queries rely on retrieving medical knowledge from large document collections. In this context, finetuning general models on medical datasets can improve retrieval relevance. Yet the dense embeddings underlying modern retrieval systems remain largely opaque, making it difficult to understand the mechanics behind their results. Moreover, adapting to a specific medical dataset risks losing transferability to new clinical corpora. In this work, we tackle these issues by applying Sparse Autoencoders (SAEs) to medical information retrieval, transforming dense text embeddings into semantically interpretable representations. To do so, our model combines a reconstruction objective with an InfoNCE contrastive loss that directly preserves query-document relevance. We evaluate three embedding models across six medical retrieval datasets, both in-domain and out-of-domain. Trained on a medical corpus, our sparse codes improve in-domain retrieval over the original embeddings, and lose less out-of-domain performance than their dense counterparts, especially when the target corpus is distant from the training one. Importantly, our experiments show that the method yields interpretable features whose semantic content can be inferred from the queries and documents that activate them most strongly. The interpretation of these features sheds light on why a particular document was retrieved, and transfers beyond the training distribution.
查看此版本解读 →查看此版本原文 ↗v2 · 2026-11-17 21:32
Explanations that Transfer: Interpreting Medical Retrieval with Sparse AutoEncoders
Medical AI systems that answer users' queries rely on retrieving medical knowledge from large document collections. In this context, finetuning general models on medical datasets can improve retrieval relevance. Yet the dense embeddings underlying modern retrieval systems remain largely opaque, making it difficult to understand the mechanics behind their results. Moreover, adapting to a specific medical dataset risks losing transferability to new clinical corpora. In this work, we tackle these issues by applying Sparse Autoencoders (SAEs) to medical information retrieval, transforming dense text embeddings into semantically interpretable representations. To do so, our model combines a reconstruction objective with an InfoNCE contrastive loss that directly preserves query-document relevance. We evaluate three embedding models across six medical retrieval datasets, both in-domain and out-of-domain. Trained on a medical corpus, our sparse codes improve in-domain retrieval over the original embeddings, and lose less out-of-domain performance than their dense counterparts, especially when the target corpus is distant from the training one. Importantly, our experiments show that the method yields interpretable features whose semantic content can be inferred from the queries and documents that activate them most strongly. The interpretation of these features sheds light on why a particular document was retrieved, and transfers beyond the training distribution.
查看此版本解读 →查看此版本原文 ↗v1 · 2026-11-13 21:48
Explanations that Transfer: Interpreting Medical Retrieval with Sparse AutoEncoders
Medical AI systems that answer users' queries rely on retrieving medical knowledge from large document collections. In this context, finetuning general models on medical datasets can improve retrieval relevance. Yet the dense embeddings underlying modern retrieval systems remain largely opaque, making it difficult to understand the mechanics behind their results. Moreover, adapting to a specific medical dataset risks losing transferability to new clinical corpora. In this work, we tackle these issues by applying Sparse Autoencoders (SAEs) to medical information retrieval, transforming dense text embeddings into semantically interpretable representations. To do so, our model combines a reconstruction objective with an InfoNCE contrastive loss that directly preserves query-document relevance. We evaluate three embedding models across six medical retrieval datasets, both in-domain and out-of-domain. Trained on a medical corpus, our sparse codes improve in-domain retrieval over the original embeddings, and lose less out-of-domain performance than their dense counterparts, especially when the target corpus is distant from the training one. Importantly, our experiments show that the method yields interpretable features whose semantic content can be inferred from the queries and documents that activate them most strongly. The interpretation of these features sheds light on why a particular document was retrieved, and transfers beyond the training distribution.
查看此版本解读 →查看此版本原文 ↗