Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Fairness Pruning: A Surgical Method to Locate Demographic Bias in LLMs

AI By Crimson AI Hugging Face Papers 2 August 2026 · 00:00 20 views
Share: X Telegram

Researchers introduce Fairness Pruning, a lightweight intervention that locates neurons responsible for demographic bias in GLU-MLP layers, showing that zeroing as few as 5 neurons can alter biased responses while preserving 99.49% of model capabilities.

Fairness Pruning: A Surgical Method to Locate Demographic Bias in LLMs

Key points

Researchers have introduced Fairness Pruning, a novel structural intervention designed to locate and manage demographic bias in large language models (LLMs). The method, detailed in a recent paper, focuses on causal bias localization using minimally contrastive prompt pairs and inference-time activation capture, identifying neurons that react differentially to demographic attributes within GLU architectures.

The approach evaluates signals at the down_proj input and has been tested on models up to 3 billion parameters, including the Llama-3.2 family and Salamandra-2B. By zeroing out identified neurons, the researchers observed that the model's responses to demographic variables change, but the effect is not a simple flat mitigation. Instead, the intervention causes bidirectional bias destabilization, as the unsigned BiasScore mixes neurons that push toward and against stereotypes, with the net effect depending on which sign dominates.

Remarkably, the intervention is extremely surgical. In Llama-3.2-1B, zeroing at most 40 neurons—less than 0.031% of the total MLP width—achieves a mean retention of 99.49% in reasoning and general knowledge capabilities. Even changing just 5 neurons is enough to significantly alter biased responses in models ranging from 1B to 3B parameters.

This work is the first in a broader research line. The next step is moving from blind zeroing to directional steering, incorporating a signed BiasScore that separates amplifier neurons from regulator neurons. All code, datasets, and results are openly available on GitHub and Hugging Face.

ModelNeurons ZeroedCapability Retention
Llama-3.2-1BUp to 4099.49%
Llama-3.2 family & Salamandra-2B5 (minimum)Significant bias alteration
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 31 Aug 2026

Hugging Face Unveils StepGuard: Step-Level Guardrails for Safer AI Agents

StepGuard, a new step-level guard model from Hugging Face, audits agent actions before execution, reducing attack success rates by...

1
Research paper
Hugging Face 31 Aug 2026

Hugging Face Researchers Unveil ABot-Recon for Stable Long-Horizon 3D Reconstruction

ABot-Recon, a new streaming 3D reconstruction model from Hugging Face, achieves stable long-horizon performance using only local t...

1
Research paper
Hugging Face 31 Aug 2026

ContextPilot: Teaching Agents Proactive Context Management via Fine-Grained RL

Hugging Face researchers introduce ContextPilot, a framework that enhances long-horizon agent reasoning by expanding context-editi...

1