---
title: "Prompt Stylometry for On-Device Affect-Adaptive AI: A Feasibility"
description: "The paper asks whether an AI assistant can read emotional and cognitive cues from how a prompt is written, not just what it… 4 references."
authors: ["Debmalya Pal"]
year: 2026
venue: "Proceedings of the Seventh Workshop on Privacy in Natural Language Processing"
source: "https://aclanthology.org/2026.privatenlp-main.6/"
doi: "10.18653/v1/2026.privatenlp-main.6"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/pal-2026-prompt-stylometry-device-affect-adaptive/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/pal-2026-prompt-stylometry-device-affect-adaptive/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/40/) / Prompt Stylometry for On-Device Affect-Adaptive AI: A Feasibility Study in Linguistic Signal Detection and Response Steering

# Prompt Stylometry for On-Device Affect-Adaptive AI: A Feasibility Study in Linguistic Signal Detection and Response Steering

**Authors:** [Debmalya Pal](https://md.research.elamur.ai/free/en/authors/debmalya-pal/) · **Year:** 2026 · **Venue:** [Proceedings of the Seventh Workshop on Privacy in Natural Language Processing](https://md.research.elamur.ai/free/en/venues/privatenlp/) ([Workshop on Privacy in Natural Language Processing 2026](https://md.research.elamur.ai/free/en/venues/privatenlp/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.privatenlp-main.6/) · [PDF](https://aclanthology.org/2026.privatenlp-main.6.pdf) · [DOI 10.18653/v1/2026.privatenlp-main.6](https://doi.org/10.18653/v1/2026.privatenlp-main.6) · [OpenAlex](https://openalex.org/W7166804910)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper asks whether an AI assistant can read emotional and cognitive cues from how a prompt is written, not just what it asks, and adapt its reply without exposing the user to psychological profiling. The author calls this "Prompt Stylometry." It covers affect cues, such as sadness, anxiety or stress rated on a 1–7 Likert scale, and cognitive-style cues: self-critical, indecisive, exploratory, analytical or confident. Server-side processing could silently profile users this way, so the whole pipeline runs locally on a consumer device. The study uses 600 synthetic prompts covering 30 stylometric profiles, generated by Llama 3.3-70B. It compares lexicon-based (VADER), neural (GoEmotions) and generative (small LLM) annotation. Response steering is done at inference time through deterministic prompt assembly, with no fine-tuning, and is tested on two small model families under 5B parameters. Annotators and model families behaved differently in systematic ways. The results suggest on-device affect-adaptive AI is feasible but immature. Annotation sensitivity and the mismatch between human and synthetic text remain open problems.

## Key points
- The paper names a privacy threat, "Implicit Profiling": any cloud LLM that handles prompts can infer users' mental states as a side effect. Fully on-device inference is proposed as the most direct countermeasure.
- The benchmark has 600 LLM-generated prompts across 30 profiles, annotated with VADER, GoEmotions and small LLMs. VADER has a single sentiment axis, so it cannot separate fine-grained negative emotions.
- The rule-based cognitive-style classifier was heavily skewed: it labelled 383 prompts analytical and 216 indecisive. The author calls this a "stylometric transfer gap," meaning psycholinguistic markers validated on human text do not carry over cleanly to LLM-generated text.
- Steering effects are hard to measure because of a floor effect. Baseline negative affect was about 1.1 on the 1–7 scale, and 84% of prompts were routed to low-distress quadrants.
- Stated limitations:
  - synthetic, single-turn data only;
  - models annotate their own inputs;
  - affect-shift is used as a proxy metric, not a measure of user benefit;
  - evaluation on a single hardware device.
- Proposed future work includes session-level tracking, IRB-approved tests with real users, and local layers that strip affect signals before prompts reach commercial APIs.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [University of California San Diego](https://md.research.elamur.ai/free/en/universities/university-of-california-san-diego/) — University, San Diego, United States · 73 works · 51,948 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 1,552 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 870 works · [💭 Social Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/social-psychology/) · 306 works

## Cite

Cite as: Debmalya Pal. 2026. Prompt Stylometry for On-Device Affect-Adaptive AI: A Feasibility Study in Linguistic Signal Detection and Response Steering. Proceedings of the Seventh Workshop on Privacy in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{pal2026prompt,
  title = {{Prompt Stylometry for On-Device Affect-Adaptive AI: A Feasibility Study in Linguistic Signal Detection and Response Steering}},
  author = {Debmalya Pal},
  booktitle = {Proceedings of the Seventh Workshop on Privacy in Natural Language Processing},
  year = {2026},
  doi = {10.18653/v1/2026.privatenlp-main.6},
  url = {https://doi.org/10.18653/v1/2026.privatenlp-main.6}
}
```

- [← previous Prompt Optimization for Relation Extraction using Reinforcement Learning](https://md.research.elamur.ai/free/en/papers/liu-2026-prompt-optimization-relation-extraction-reinforcement/)
- [next → Propaganda Signals in LLMs: Perspectival Divergence and Narrative Framing in the Russia-Ukraine War](https://md.research.elamur.ai/free/en/papers/shabat-2026-propaganda-signals-llms-perspectival-divergence/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/40/)

## References

4 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Meta AI. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). Preprint, arXiv:2407.21783.
2. Mohammed Al-Mosaiwi and Tom Johnstone. 2018. In an absolute state: Elevated use of absolutist words is a marker specific to anxiety, depression, and suicidal ideation. Clinical Psychological Science, 6(4):529– 542. [find](https://scholar.google.com/scholar?q=Mohammed+Al%2DMosaiwi+and+Tom+Johnstone%2E+2018%2E+In+an+absolute+state%3A+Elevated+use+of+absolutist+words+is+a+marker+specific+to+anxiety%2C+depression%2C+and+suicidal+ideation%2E+Clinical+Psychological+Science%2C+)
3. Aaron T. Beck, A. John Rush, Brian F. Shaw, and Gary Emery. 1979. Cognitive Therapy of Depression. Guilford Press, New York. [find](https://scholar.google.com/scholar?q=Aaron+T%2E+Beck%2C+A%2E+John+Rush%2C+Brian+F%2E+Shaw%2C+and+Gary+Emery%2E+1979%2E+Cognitive+Therapy+of+Depression%2E+Guilford+Press%2C+New+York%2E)
4. Hannah Brown, Katherine Lee, Fatemehsadat Mireshghallah, Reza Shokri, and Florian Tramèr. 2022. What does it mean for a language model to preserve privacy? In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency. \#\# Conclusion [find](https://scholar.google.com/scholar?q=Hannah+Brown%2C+Katherine+Lee%2C+Fatemehsadat+Mireshghallah%2C+Reza+Shokri%2C+and+Florian+Trame%CC%80r%2E+2022%2E+What+does+it+mean+for+a+language+model+to+preserve+privacy%3F+In+Proceedings+of+the+2022+ACM+Conference+o)

*Updated 1 November 2026*