# The Assistant as a Privileged Persona: A canonical reference in cross-persona self-recognition

arXiv: https://arxiv.org/abs/2606.00545 · HTML: https://mlsanity.com/paper/2606.00545/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 18 references that are in the map.
- First posted on arXiv 30 May 2026.

## Builds on (most cited first, 16 of 16)

- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [An Explanation of In-context Learning as Implicit Bayesian Inference](https://mlsanity.com/paper/2111.02080/) (2021)
- [DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature](https://mlsanity.com/paper/2301.11305/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Understanding the Effects of RLHF on LLM Generalisation and Diversity](https://mlsanity.com/paper/2310.06452/) (2023)
- [Role-Play with Large Language Models](https://mlsanity.com/paper/2305.16367/) (2023)
- [Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text](https://mlsanity.com/paper/2401.12070/) (2024)
- [Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning](https://mlsanity.com/paper/2301.11916/) (2023)
- [Language Models as Agent Models](https://mlsanity.com/paper/2212.01681/) (2022)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Tell me about yourself: LLMs are aware of their learned behaviors](https://mlsanity.com/paper/2501.11120/) (2025)
- [Creativity Has Left the Chat: The Price of Debiasing Language Models](https://mlsanity.com/paper/2406.05587/) (2024)
- [Self-Recognition in Language Models](https://mlsanity.com/paper/2407.06946/) (2024)
- [Inspection and Control of Self-Generated-Text Recognition Ability in Llama3-8b-Instruct](https://mlsanity.com/paper/2410.02064/) (2024)
- [From Simulation to Enaction: Post-trained language models recognize and react to their own generations](https://mlsanity.com/paper/2605.25459/) (2026)

## Abstract (arXiv)

Post-trained language models can recognize their own outputs from a sentence or two out of context. In a companion paper citepjack2026twomodes we showed they can also recognize when they are currently acting on-policy, through the sharp entropy drop of assistant-mode generation. Both signals are tied to the Assistant persona that post-training mainly shapes. This paper widens the frame to cross-persona authorship judgement on Llama-3.1-70B-Instruct. We measure a matrix of authorship claim rates over a panel of evaluator and generator personas spanning librarian to dragon to Shakespeare, and make two claims. First, on the Assistant's own row of the matrix, the Assistant's claim rate, the persona-vector distance from the Assistant in activation space, and the entropy gap between the Assistant's surprise on a persona's text and the persona's surprise on its own text are all tightly coupled. This extends the entropy signature of acting from the companion paper to a retrospective signature of having acted. Second, this coupling fails off the Assistant's row: the natural symmetric extension of the entropy gap does not predict authorship for distinctive evaluators (pirate, dragon, Shakespeare); what does is asymmetric -- the evaluator's surprise compared to the Assistant's surprise on the same text, not to the generator's. We rule out the alternative that any persona could play this reference role by trying many candidate substitutes; none does. We interpret the asymmetry as the model performing an implicit Bayesian likelihood-ratio test against the Assistant as the canonical alternative hypothesis, with the persona-vector geometry of citetchen2025persona (every persona a delta off the Assistant) ensuring that the Assistant is the only persona universally accessible to that test.