Pith. sign in

REVIEW 4 cited by

Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.03010 v1 pith:6ZEHHRKQ submitted 2023-05-04 cs.CL cs.CR

classification cs.CLcs.CR
keywords embeddingsentenceinversionrepresentationsattackgenerativeinformationlanguage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Sentence-level representations are beneficial for various natural language processing tasks. It is commonly believed that vector representations can capture rich linguistic properties. Currently, large language models (LMs) achieve state-of-the-art performance on sentence embedding. However, some recent works suggest that vector representations from LMs can cause information leakage. In this work, we further investigate the information leakage issue and propose a generative embedding inversion attack (GEIA) that aims to reconstruct input sequences based only on their sentence embeddings. Given the black-box access to a language model, we treat sentence embeddings as initial tokens' representations and train or fine-tune a powerful decoder model to decode the whole sequences directly. We conduct extensive experiments to demonstrate that our generative inversion attack outperforms previous embedding inversion attacks in classification metrics and generates coherent and contextually similar sentences as the original inputs.

Discussion (0). Sign in to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. An Attack to Break Permutation-Based Private Third-Party Inference Schemes for LLMs

    cs.CR 2025-05 conditional novelty 8.0 of 10

    A sequential vocabulary-search attack decodes original prompts from unpermuted and permuted LLM hidden states, compromising PermLLM, STIP, and Centaur.

  2. inversedMixup: Data Augmentation via Inverting Mixed Embeddings

    cs.CL 2026-01 conditional novelty 6.0 of 10

    Mixing BERT embeddings and inverting them into text with LLaMA produces interpretable augmented sentences, improves few-shot classification on some datasets, and exposes 'manifold intrusion' in text Mixup.

  3. Depth Gives a False Sense of Privacy: LLM Internal States Inversion

    cs.CR 2025-07 conditional novelty 6.0 of 10

    LLM internal states at intermediate layers contain enough information to recover long, sensitive user prompts with high accuracy.

  4. Transforming Sensitive Documents into Quantitative Data: An AI-Based Preprocessing Toolchain for Structured and Privacy-Conscious Analysis

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A local open-weight LLM toolchain transforms sensitive Swedish court decisions into anonymized English summaries and embeddings that retain semantic content, validated on 10,842 LVM documents and a suicide-related con...

Pith tools