proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersYutong Song, Jiang Wu, Shaofan Yuan1 min readpaperadvanced

Do Implicit Personalization and Explicit Styles Conflict? PsPLUG: A Lightweight Plug-in for Balancing Personalization and Style in Customized LLMs

Read on arxiv.org

Summary

Personalized LLMs often lose user-specific characteristics when explicit style instructions are applied, a problem called personalization collapse. PsPLUG is a lightweight plug-in that addresses this by learning a user-specific residual, allowing dynamic control over personalization strength.

  • Explicit style instructions can cause "personalization collapse" in LLMs, diminishing user-specific traits.
  • PsPLUG is a lightweight plug-in designed to mitigate this conflict between personalization and style.
  • It works by learning a user-specific residual after accounting for the requested style.
  • PsPLUG enables tuning personalization strength at inference time, offering fine-grained control.

Engineers building customized LLMs need to understand how to balance user personalization with explicit style adherence, which PsPLUG offers a novel approach to.

7/10
llmmachine learningainlppersonalizationmodel tuning

Related reading

  1. Persona Dosing: Calibrated Activation Steering for Graded Trait Control

    PersonaDose trains a description‑conditioned FLAS controller to steer LLM persona traits and calibrates flow time to hit a requested intensity, without needing intensity‑labeled data. Across Llama‑3.1‑8B, Qwen3‑8B, and Gemma‑3‑4B it boosts trait expression by up to 33 points and achieves mean targeting errors of 4.7‑6.2 points over reachable targets.

    Hugging Face Daily Papersarxiv.org1 minpaper

  4. DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation

    DeformSmith is a framework that generates physically plausible deformable assets for robot manipulation from a text prompt or a single image, using a hierarchical construction process guided by a shared physics harness. It outperforms prior baselines in visual fidelity and physical realism while also producing interaction data for downstream tasks.

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. Article: Beyond Relevance: A Governance-First Architecture for Enterprise Personalization

    The article proposes a governance‑first architecture for enterprise personalization, where policy‑driven steps (memory, journey graph, AI routing, scoring, trust checks, outcome simulation) shape the recommendation before it is returned. A reference FastAPI implementation demonstrates the pattern with external YAML policies and optional LLM assistance.

    InfoQinfoq.com19 min