Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Personalized Privacy Control in LLMs via Attention Head Intervention

Junseok Kim, Nakyeong Yang, Kyomin Jung

Published

Aug 21, 2026

Citations

0

Trust level

Low

Usefulness score

40/100 (Low)

Extraction confidence

45% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Aug 21, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

General Pairwise Preference

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this as background context only. Do not make protocol decisions from this page alone.

Best use

Background context only

Use if you need

Background context only.

What to verify

Read the full paper before copying any benchmark, metric, or protocol choices.

Main weakness

The available metadata is too thin to trust this as a primary source.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Weak or implicit
Validate from full paper
Usefulness for eval research
40/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns. Prior work on contextual privacy studies whether LLMs regulate information disclosure according to context-dependent norms. However, acceptable disclosure boundaries may vary across users even within the same context. To address this limitation, we introduce \textit{personalized privacy}, which incorporates user-specific disclosure preferences into privacy control. We further present P3Bench~(\textbf{P}ersonalized \textbf{P}rivacy \textbf{P}reservation \textbf{Bench}mark), a novel benchmark extending contextual privacy policies with personalized disclosure policies. Experiments show that prompt-based policies fail to reliably enforce personalized privacy policies, with Qwen2.5-7B and Gemma3-4B showing average policy ignorance ratios of 51.25\% and 74.28\%, respectively. Finally, to address this problem, we propose \textsc{Repair}, a robust inference-time attention head intervention method that adjusts disclosure behavior toward policy-consistent responses. Our method significantly improves adherence to user-specific privacy preferences by reducing cases where the model fails to follow the given policy.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

partial

Pairwise Preference

Directly usable for protocol triage.

"The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns."

Evaluation Modes

missing

None explicit

Validate eval design from full paper text.

"The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns."

Quality Controls

missing

Not reported

No explicit QC controls found.

"The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns."

Reported Metrics

missing

Not extracted

No metric anchors detected.

"The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns."

Benchmarks and datasets

No benchmark or dataset names were extracted from the available abstract.

Reported metrics

No metric terms were extracted from the available abstract.

Human feedback details
Uses human feedback
Yes
Feedback types
Pairwise Preference
Rater population
Not reported
Expertise required
General
Evaluation details
Evaluation modes
None
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns.
  • Prior work on contextual privacy studies whether LLMs regulate information disclosure according to context-dependent norms.
  • However, acceptable disclosure boundaries may vary across users even within the same context.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Contribution summary

  • The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns.
  • To address this limitation, we introduce personalized privacy, which incorporates user-specific disclosure preferences into privacy control.
  • Finally, to address this problem, we propose Repair, a robust inference-time attention head intervention method that adjusts disclosure behavior toward policy-consistent responses.

Why it matters for eval

  • The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns.
  • To address this limitation, we introduce personalized privacy, which incorporates user-specific disclosure preferences into privacy control.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Pairwise Preference

  • Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Metric reporting is present

    No metric terms extracted.