# Aggretriever: A Simple Approach to Aggregate Textual Representations for Robust Dense Passage Retrieval

arXiv: https://arxiv.org/abs/2208.00511 · HTML: https://mlsanity.com/paper/2208.00511/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2022.

## Built on by (most cited first, 12 of 12)

- [Mitigating Object Hallucination in Large Vision-Language Models via Image-Grounded Guidance](https://mlsanity.com/paper/2402.08680/) (2024)
- [LED: Lexicon-Enlightened Dense Retriever for Large-Scale Retrieval](https://mlsanity.com/paper/2208.13661/) (2022)
- [KTRL+F: Knowledge-Augmented In-Document Search](https://mlsanity.com/paper/2311.08329/) (2023)
- [Synergistic Approach for Simultaneous Optimization of Monolingual, Cross-lingual, and Multilingual Information Retrieval](https://mlsanity.com/paper/2408.10536/) (2024)
- [CRAFT: Training-Free Cascaded Retrieval for Tabular QA](https://mlsanity.com/paper/2505.14984/) (2025)
- [Delta-KNN: Improving Demonstration Selection in In-Context Learning for Alzheimer's Disease Detection](https://mlsanity.com/paper/2506.03476/) (2025)
- [Completing Missing Annotation: Multi-Agent Debate for Accurate and Scalable Relevant Assessment for IR Benchmarks](https://mlsanity.com/paper/2602.06526/) (2026)
- [Domain Adaptation of Multilingual Semantic Search -- Literature Review](https://mlsanity.com/paper/2402.02932/) (2024)
- [Semi-Parametric Retrieval via Binary Bag-of-Tokens Index](https://mlsanity.com/paper/2405.01924/) (2024)
- [GOLFer: Smaller LM-Generated Documents Hallucination Filter & Combiner for Query Expansion in Information Retrieval](https://mlsanity.com/paper/2506.04762/) (2025)
- [Cross-Sensory Brain Passage Retrieval: Scaling Beyond Visual to Audio](https://mlsanity.com/paper/2601.14001/) (2026)
- [Baszta: Data-Centric Fine-Tuning of a Polish Multi-Label Safety Classifier](https://mlsanity.com/paper/2609.29266/) (2026)

## Abstract (arXiv)

Pre-trained language models have been successful in many knowledge-intensive NLP tasks. However, recent work has shown that models such as BERT are not ``structurally ready'' to aggregate textual information into a [CLS] vector for dense passage retrieval (DPR). This ``lack of readiness'' results from the gap between language model pre-training and DPR fine-tuning. Previous solutions call for computationally expensive techniques such as hard negative mining, cross-encoder distillation, and further pre-training to learn a robust DPR model. In this work, we instead propose to fully exploit knowledge in a pre-trained language model for DPR by aggregating the contextualized token embeddings into a dense vector, which we call agg*. By concatenating vectors from the [CLS] token and agg*, our Aggretriever model substantially improves the effectiveness of dense retrieval models on both in-domain and zero-shot evaluations without introducing substantial training overhead. Code is available at https://github.com/castorini/dhr