Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models

Binchi Zhang, Xujiang Zhao, Jundong Li, Haifeng Chen, Zhengzhang Chen

TL;DR

CultureManager synthesizes task-aware cultural data in line with target task formats, grounded in culturally relevant web search results, and manages multi-culture knowledge learned in separate adapters with a culture router that selects the appropriate one to apply.

Abstract

Large language models (LLMs) are increasingly deployed in culturally sensitive real-world tasks. However, existing cultural alignment approaches fail to align LLMs' broad cultural values with the specific goals of downstream tasks and suffer from cross-culture interference. We propose CultureManager, a novel pipeline for task-specific cultural alignment. CultureManager synthesizes task-aware cultural data in line with target task formats, grounded in culturally relevant web search results. To prevent conflicts between cultural norms, it manages multi-culture knowledge learned in separate adapters with a culture router that selects the appropriate one to apply. Experiments across ten national cultures and culture-sensitive tasks show consistent improvements over prompt-based and fine-tuning baselines. Our results demonstrate the necessity of task adaptation and modular culture management for effective cultural alignment.

Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models

TL;DR

CultureManager synthesizes task-aware cultural data in line with target task formats, grounded in culturally relevant web search results, and manages multi-culture knowledge learned in separate adapters with a culture router that selects the appropriate one to apply.

Abstract

Large language models (LLMs) are increasingly deployed in culturally sensitive real-world tasks. However, existing cultural alignment approaches fail to align LLMs' broad cultural values with the specific goals of downstream tasks and suffer from cross-culture interference. We propose CultureManager, a novel pipeline for task-specific cultural alignment. CultureManager synthesizes task-aware cultural data in line with target task formats, grounded in culturally relevant web search results. To prevent conflicts between cultural norms, it manages multi-culture knowledge learned in separate adapters with a culture router that selects the appropriate one to apply. Experiments across ten national cultures and culture-sensitive tasks show consistent improvements over prompt-based and fine-tuning baselines. Our results demonstrate the necessity of task adaptation and modular culture management for effective cultural alignment.
Paper Structure (18 sections, 1 equation, 4 figures, 3 tables)

This paper contains 18 sections, 1 equation, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Preliminary
  5. Search Query Generation
  6. Data Synthesis
  7. Cultural Alignment and Management
  8. Experiments
  9. Datasets
  10. Baselines
  11. Main Results
  12. Ablation Study
  13. Data Synthesis Examples
  14. Culture Router Utility
  15. Task Adaptation
  16. ...and 3 more sections

Figures (4)

  • Figure 1: Overview of the CultureManager pipeline. CultureManager consists of three parts: Search Query Generation (left), Task-aware Cultural Data Synthesis (center), and Culture Management (right).
  • Figure 2: Experimental results of the ablation study. "specific-only": use only task-specific queries in data synthesis; "agnostic-only": use only task-agnostic queries in data synthesis; "all-culture": abandon knowledge management and train the model on all datasets.
  • Figure 3: Accuracy of the culture router on different culture-sensitive tasks.
  • Figure 4: Results of task adaptation on different cultures. SFT represents CultureSFT; SFT+Adaptation means firstly fine-tuning the LLM using CultureSFT and then continually fine-tuning it using the synthetic data; TaskSFT denotes directly fine-tuning the original LLM using the synthetic data. The metric of each culture is computed as the mean value across the tasks under that culture.