Compass-Embedding v4: Robust Contrastive Learning for Multilingual E-commerce Embeddings

TL;DR

Compass-Embedding v4 introduces a robust, multilingual embedding framework optimized for Southeast Asian e-commerce, addressing low-resource language challenges, false negatives in contrastive learning, and deployment efficiency.

Contribution

It proposes Class-Aware Masking for improved contrastive learning, diversified synthetic data generation for low-resource languages, and deployment optimizations like spherical model merging and FP8 quantization.

Findings

State-of-the-art performance on SEA languages

Significant improvements in domain-specific retrieval

Effective handling of low-resource language data

Abstract

As global e-commerce rapidly expands into emerging markets, the lack of high-quality semantic representations for low-resource languages has become a decisive bottleneck for retrieval, recommendation, and search systems. In this work, we present Compass-Embedding v4, a high-efficiency multilingual embedding framework specifically optimized for Southeast Asian (SEA) e-commerce scenarios, where data scarcity, noisy supervision, and strict production constraints jointly challenge representation learning. Compass-Embedding v4 addresses three core challenges. First, large-batch contrastive training under mixed task supervision introduces systematic false negatives that degrade semantic alignment. We propose Class-Aware Masking (CAM), a lightweight modification to the InfoNCE objective that suppresses invalid in-batch negatives and improves semantic discrimination without altering training efficiency. Second, low-resource SEA languages suffer from limited and uneven data coverage. We construct a diversified training corpus through context-grounded synthetic data generation, cross-lingual translation, and structured e-commerce data construction, enabling robust multilingual and domain-specific learning. Third, production deployment requires high-throughput inference while preserving embedding quality. We combine robustness-driven large-batch training with spherical model merging to mitigate catastrophic forgetting, and optimize inference via vLLM and FP8 quantization. Extensive evaluations across multilingual benchmarks and proprietary e-commerce tasks show that Compass-Embedding v4 achieves state-of-the-art performance on major SEA languages, significantly outperforming general-purpose embedding models in domain-specific retrieval and classification, while maintaining competitive performance on high-resource languages.