Wen-tau Yih · page 2

29 papers in our library · 2004–2025 · page 2 of 2

Papers

The most cited first (OpenAlex and our library).

How to Train Your Dragon: Diverse Augmentation Towards Generalizable Dense Retrieval (2023 · 116 citations)
Dense retrievers (DR) of BERT-base size have usually forced a choice: models tuned for in-domain supervised retrieval on MS MARCO tend to transfer poorly to new domains, and the reverse also holds.
Retrieval-Augmented Multimodal Language Modeling (2022 · 104 citations)
Large multimodal generators such as DALL-E and CM3 keep all of their world knowledge inside their weights.
Reliable, Adaptable, and Attributable Language Models with Retrieval (2024 · 45 citations)
This position paper by Asai et al. contends that retrieval-augmented LMs should succeed purely parametric LMs as the main design for future language models. It contains no new experiments, datasets, or models. Instead, it surveys prior work and lists five weaknesses of parametric models: hallucinated facts, outputs…
DRAMA: Diverse Augmentation from Large Language Models to Smaller Dense Retrievers (2025 · 13 citations)
Billion-parameter LLMs fine-tuned as dense retrievers generalize well but are expensive at inference. For example, an 8B Llama backbone costs roughly 40× more than a BERT-based retriever. Small encoder models are cheap but transfer poorly when supervised data is scarce. DRAMA is a framework that uses LLMs to build…