Papers
arxiv:2602.16835

NeST: Neuron Selective Tuning for LLM Safety

Published on Feb 18
· Submitted by
Lichao Wu
on Feb 20
Authors:
,
,
,

Abstract

NeST is a lightweight safety alignment framework that selectively adapts safety-relevant neurons while keeping the rest of the model frozen, achieving significant reductions in unsafe generations with minimal trainable parameters.

AI-generated summary

Safety alignment is essential for the responsible deployment of large language models (LLMs). Yet, existing approaches often rely on heavyweight fine-tuning that is costly to update, audit, and maintain across model families. Full fine-tuning incurs substantial computational and storage overhead, while parameter-efficient methods such as LoRA trade efficiency for inconsistent safety gains and sensitivity to design choices. Safety intervention mechanisms such as circuit breakers reduce unsafe outputs without modifying model weights, but do not directly shape or preserve the internal representations that govern safety behavior. These limitations hinder rapid and reliable safety updates, particularly in settings where models evolve frequently or must adapt to new policies and domains. We present NeST, a lightweight, structure-aware safety alignment framework that strengthens refusal behavior by selectively adapting a small subset of safety-relevant neurons while freezing the remainder of the model. NeST aligns parameter updates with the internal organization of safety behavior by clustering functionally coherent safety neurons and enforcing shared updates within each cluster, enabling targeted and stable safety adaptation without broad model modification or inference-time overhead. We benchmark NeST against three dominant baselines: full fine-tuning, LoRA-based fine-tuning, and circuit breakers across 10 open-weight LLMs spanning multiple model families and sizes. Across all evaluated models, NeST reduces the attack success rate from an average of 44.5% to 4.36%, corresponding to a 90.2% reduction in unsafe generations, while requiring only 0.44 million trainable parameters on average. This amounts to a 17,310x decrease in updated parameters compared to full fine-tuning and a 9.25x reduction relative to LoRA, while consistently achieving stronger safety performance for alignment.

Community

We introduce NeST, a lightweight and structure-aware safety alignment framework that selectively adapts clustered safety-relevant neurons to deliver stable, efficient, and model-agnostic refusal strengthening—achieving a 90.2% reduction in unsafe generations while updating only 0.44M parameters across 10 open-weight LLMs.

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2602.16835 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2602.16835 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2602.16835 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.