# PTP: Previous-Token Prediction based LLM Inversion for Near-Exact Prompt Reconstruction

arXiv: https://arxiv.org/abs/2607.29378 · HTML: https://mlsanity.com/paper/2607.29378/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 31 Jul 2026.

## Builds on (most cited first, 12 of 12)

- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Text Embeddings Reveal (Almost) As Much As Text](https://mlsanity.com/paper/2310.06816/) (2023)
- [Effective Prompt Extraction from Language Models](https://mlsanity.com/paper/2307.06865/) (2023)
- [Model Inversion Networks for Model-Based Optimization](https://mlsanity.com/paper/1912.13464/) (2019)
- [Language Model Inversion](https://mlsanity.com/paper/2311.13647/) (2023)
- [Extracting Prompts by Inverting LLM Outputs](https://mlsanity.com/paper/2405.15012/) (2024)
- [Autoinverse: Uncertainty Aware Inversion of Neural Networks](https://mlsanity.com/paper/2208.13780/) (2022)
- [Landscape Learning for Neural Network Inversion](https://mlsanity.com/paper/2206.09027/) (2022)
- [Network Inversion of Binarised Neural Nets](https://mlsanity.com/paper/2402.11995/) (2024)
- [Better Language Model Inversion by Compactly Representing Next-Token Distributions](https://mlsanity.com/paper/2506.17090/) (2025)
- [Network Inversion of Convolutional Neural Nets](https://mlsanity.com/paper/2407.18002/) (2024)
- [Reverse Prompt Engineering](https://mlsanity.com/paper/2411.06729/) (2024)

## Abstract (arXiv)

Large language models (LLMs) generate text by auto-regressively sampling the next token. This inherently leads to a many-to-many mapping between prompts and responses, complicating the task of inferring prompts from observed outputs. Prior work on LLM inversion frames prompt recovery as a semantic reconstruction task. They rely on fine-tuning pretrained sequence-to-sequence models on large external datasets--and requiring access to model weights or logits--to generate semantically plausible prompts. In contrast, we present a functional approach to inverting a given LLM in a black-box setting, without auxiliary aids. We train an explicit inverse language model entirely from scratch on data synthetically generated from the target LLM itself. Analogous to forward next-token prediction, our inverse model is trained using previous-token prediction, establishing a generative link between the forward and inverse processes that enables faithful prompt reconstruction. Moreover, it naturally supports diverse prompt reconstructions through sampling, whereby all such prompts induce similar responses under the forward, target LLM. Our approach generalises across datasets and exhibits transferability in reconstructing prompts from responses generated by different LLMs. Further, across the set of token based evaluation metrics for prompt and response reconstructions, our approach outperforms prior work.