# Multilingual Fine-Tuning via Localized Gradient Conflict Resolution

arXiv: https://arxiv.org/abs/2606.05613 · HTML: https://mlsanity.com/paper/2606.05613/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 43 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Builds on (most cited first, 30 of 35)

- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [Multi-Task Learning as Multi-Objective Optimization](https://mlsanity.com/paper/1810.04650/) (2018)
- [PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel](https://mlsanity.com/paper/2304.11277/) (2023)
- [Conflict-Averse Gradient Descent for Multi-task Learning](https://mlsanity.com/paper/2110.14048/) (2021)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications](https://mlsanity.com/paper/2402.05162/) (2024)
- [Multi-Task Learning as a Bargaining Game](https://mlsanity.com/paper/2202.01017/) (2022)
- [Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout](https://mlsanity.com/paper/2010.06808/) (2020)
- [Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation](https://mlsanity.com/paper/2412.03304/) (2024)
- [Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models](https://mlsanity.com/paper/2010.05874/) (2020)
- [How do Large Language Models Handle Multilingualism?](https://mlsanity.com/paper/2402.18815/) (2024)
- [The Geometry of Multilingual Language Model Representations](https://mlsanity.com/paper/2205.10964/) (2022)
- [MetricX-24: The Google Submission to the WMT 2024 Metrics Shared Task](https://mlsanity.com/paper/2410.03983/) (2024)
- [WMT24++: Expanding the Language Coverage of WMT24 to 55 Languages & Dialects](https://mlsanity.com/paper/2502.12404/) (2025)
- [Monolingual or Multilingual Instruction Tuning: Which Makes a Better Alpaca](https://mlsanity.com/paper/2309.08958/) (2023)
- [PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts](https://mlsanity.com/paper/2504.18428/) (2025)
- [Recon: Reducing Conflicting Gradients from the Root for Multi-Task Learning](https://mlsanity.com/paper/2302.11289/) (2023)
- [Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languages](https://mlsanity.com/paper/2402.12204/) (2024)
- [G-DIG: Towards Gradient-based Diverse and High-quality Instruction Data Selection for Machine Translation](https://mlsanity.com/paper/2405.12915/) (2024)
- [Multilingual Routing in Mixture-of-Experts](https://mlsanity.com/paper/2510.04694/) (2025)
- [Towards Understanding Multi-Task Learning (Generalization) of LLMs via Detecting and Exploring Task-Specific Neurons](https://mlsanity.com/paper/2407.06488/) (2024)
- [The Emergence of Abstract Thought in Large Language Models Beyond Any Language](https://mlsanity.com/paper/2506.09890/) (2025)
- [SEA-HELM: Southeast Asian Holistic Evaluation of Language Models](https://mlsanity.com/paper/2502.14301/) (2025)
- [When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners](https://mlsanity.com/paper/2505.15257/) (2025)

## Abstract (arXiv)

Cross-lingual versatility has become a defining capability of Large Language Models (LLMs). However, fine-tuning these models often induces negative interference, where gains in one language come at another's expense. To address this, we cast multilingual fine-tuning as a multi-objective optimization (MOO) problem. Specifically, we introduce Bucket-Level MOO, a scalable distributed framework that applies gradient-based MOO algorithms locally, within the parameter buckets of distributed training. This enables conflict-aware updates without ever materializing a full gradient vector for any objective, and improves multilingual performance over standard fine-tuning by up to 4.3 points. Theoretically, we define bucketwise Pareto stationarity, show it is necessary for Pareto optimality under any fixed partition, and prove descent and convergence for representative Bucket-Level variants. Our work brings conventional MOO within reach for multilingual LLMs at a far lower cost. More broadly, Bucket-Level MOO is agnostic to both the objectives and the update rule, readily extending to other multi-task learning at scale.