# Exploring the Representation Power of SPLADE Models

arXiv: https://arxiv.org/abs/2306.16680 · HTML: https://mlsanity.com/paper/2306.16680/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jun 2023.

## Built on by (most cited first, 5 of 5)

- [LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum](https://mlsanity.com/paper/2601.01684/) (2026)
- [The Role of Vocabularies in Learning Sparse Representations for Ranking](https://mlsanity.com/paper/2509.16621/) (2025)
- [LightRetriever: A LLM-based Text Retrieval Architecture with Extremely Faster Query Inference](https://mlsanity.com/paper/2505.12260/) (2025)
- [The Pre-Training Study of Expanded-SPLADE Models on Web Document Titles](https://mlsanity.com/paper/2605.01407/) (2026)
- [Rescaling MLM-Head for Neural Sparse Retrieval](https://mlsanity.com/paper/2606.18811/) (2026)

## Abstract (arXiv)

The SPLADE (SParse Lexical AnD Expansion) model is a highly effective approach to learned sparse retrieval, where documents are represented by term impact scores derived from large language models. During training, SPLADE applies regularization to ensure postings lists are kept sparse -- with the aim of mimicking the properties of natural term distributions -- allowing efficient and effective lexical matching and ranking. However, we hypothesize that SPLADE may encode additional signals into common postings lists to further improve effectiveness. To explore this idea, we perform a number of empirical analyses where we re-train SPLADE with different, controlled vocabularies and measure how effective it is at ranking passages. Our findings suggest that SPLADE can effectively encode useful ranking signals in documents even when the vocabulary is constrained to terms that are not traditionally useful for ranking, such as stopwords or even random words.