paper-with-me

Papers

CONGRAD:Conflicting Gradient Filtering for Multilingual Preference Alignment

2025-03-31 · Jiangnan Li, Thuy-Trang Vu, Christian Herold, Amirhossein Tebbifakhr, Shahram Khadivi, Gholamreza Haffari

Naive joint training of large language models (LLMs) for multilingual preference alignment can suffer from negative interference. This is a known issue in multilingual training, where conflicting objectives degrade overall performance. However, the impact of this phenomenon in the context of multilingual preference alignment remains largely underexplored. To address this issue, we propose CONGRAD, a scalable and effective filtering method that selects high-quality preference samples with minimal gradient conflicts across languages. Our method leverages gradient surgery to retain samples aligned with an aggregated multilingual update direction. Additionally, we incorporate a sublinear gradient compression strategy that reduces memory overhead during gradient accumulation. We integrate CONGRAD into self-rewarding framework and evaluate on LLaMA3-8B and Gemma2-2B across 10 languages. Results show that CONGRAD consistently outperforms strong baselines in both seen and unseen languages, with minimal alignment tax.

📄 PDF Abstract BibTeX arXiv:2503.23777

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Language Bias under Conflicting Information in Multilingual LLMs

2026-04-08 · Robert Östling, Murathan Kurfalı arxiv

Large Language Models (LLMs) have been shown to contain biases in the process of integrating conflicting information when answering questions. Here we ask whether such biases also exist with respect to which language is …

Drinking from a Firehose: Continual Learning with Web-scale Natural Language

2020-07-18 · Hexiang Hu, Ozan Sener, Fei Sha, Vladlen Koltun

Continual learning systems will interact with humans, with each other, and with the physical world through time -- and continue to learn and adapt as they do. An important open problem for continual learning is a large-s…

Continual Learning

Gradient Deconfliction via Orthogonal Projections onto Subspaces For Multi-task Learning

2025-03-05 · Shijie Zhu, Hui Zhao, Tianshu Wu, Pengjie Wang 외

Although multi-task learning (MTL) has been a preferred approach and successfully applied in many real-world scenarios, MTL models are not guaranteed to outperform single-task models on all tasks mainly due to the negati…

Multi-Task Learning

Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization

2026-04-27 · Xinxin Liu, Ming Li, Zonglin Lyu, Yuzhang Shang 외 arxiv

Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide only single, holistic annotations, resulting in severe label…

Reward-free Alignment for Conflicting Objectives

2026-02-02 · Peter Chen, Xiaopeng Li, Xi Chen, Tianyi Lin arxiv

Direct alignment methods are increasingly used to align large language models (LLMs) with human preferences. However, many real-world alignment problems involve multiple conflicting objectives, where naive aggregation of…