# Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings

arXiv: https://arxiv.org/abs/2606.07502 · HTML: https://mlsanity.com/paper/2606.07502/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 33 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Builds on (most cited first, 22 of 22)

- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models](https://mlsanity.com/paper/2104.08663/) (2021)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings](https://mlsanity.com/paper/1909.00512/) (2019)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [Fact or Fiction: Verifying Scientific Claims](https://mlsanity.com/paper/2004.14974/) (2020)
- [SPECTER: Document-level Representation Learning using Citation-informed Transformers](https://mlsanity.com/paper/2004.07180/) (2020)
- [LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders](https://mlsanity.com/paper/2404.05961/) (2024)
- [Matryoshka Representation Learning](https://mlsanity.com/paper/2205.13147/) (2022)
- [Unsupervised Corpus Aware Language Model Pre-training for Dense Passage Retrieval](https://mlsanity.com/paper/2108.05540/) (2021)
- [Whitening Sentence Representations for Better Semantics and Faster Retrieval](https://mlsanity.com/paper/2103.15316/) (2021)
- [Scaling Sentence Embeddings with Large Language Models](https://mlsanity.com/paper/2307.16645/) (2023)
- [On the Sentence Embeddings from Pre-trained Language Models](https://mlsanity.com/paper/2011.05864/) (2020)
- [Spectral Filters, Dark Signals, and Attention Sinks](https://mlsanity.com/paper/2402.09221/) (2024)
- [Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free](https://mlsanity.com/paper/2410.10814/) (2024)
- [Searching for Scientific Evidence in a Pandemic: An Overview of TREC-COVID](https://mlsanity.com/paper/2104.09632/) (2021)
- [Meta-Task Prompting Elicits Embeddings from Large Language Models](https://mlsanity.com/paper/2402.18458/) (2024)
- [GenEOL: Harnessing the Generative Power of LLMs for Training-Free Sentence Embeddings](https://mlsanity.com/paper/2410.14635/) (2024)
- [A Text is Worth Several Tokens: Text Embedding from LLMs Secretly Aligns Well with The Key Tokens](https://mlsanity.com/paper/2406.17378/) (2024)

## Built on by (most cited first, 1 of 1)

- [Masking Frequent Tokens Sharpens Direct Preference Optimization](https://mlsanity.com/paper/2609.32445/) (2026)

## Abstract (arXiv)

Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlying this deficiency. Our motivation stems from an unexpected observation: text embeddings tend to align with frequent but uninformative tokens when projected onto the vocabulary space. We argue that this excessive expression of high-frequency tokens suppresses the model's ability to capture nuanced semantics. To address this, we introduce EmbedFilter, a simple linear transformation designed to refine text embeddings derived from LLMs directly. Specifically, we uncover that the unembedding matrix within LLMs encodes a latent space that is actively writing these frequent tokens into embedding space. By filtering out this subspace, EmbedFilter suppress the influence of high-frequency tokens, thereby enhancing semantic representations. As a compelling byproduct, this enables an inherent dimensionality reduction, lowering index storage and speedup retrieval while fully preserving the refined embedding quality. Our experiments across multiple LLM backbones demonstrate that LLMs equipped with EmbedFilter achieve superior zero-shot downstream performance even with significantly reduced embedding dimensions. We hope our findings provide deeper insights into the mechanisms of LLM-based representations and inspire more principled designs to improve text embeddings training. Our code is available at https://github.com/CentreChen/EmbFilter.