5 papers
SARA: Unlocking Multilingual Knowledge in Mixture-of-Experts via Semantically Anchored Routing Alignment
Tianyu Dong, Yangyang Liu, Jiang Zhou +9
Sparse Mixture-of-Experts (MoE) architectures have emerged as an increasingly influential paradigm as they offer a strategic balance between parameter scalability and computational…
Incentivizing Parametric Knowledge via Reinforcement Learning with Verifiable Rewards for Cross-Cultural Entity Translation
Jiang Zhou, Xiaohu Zhao, Xinwei Wu +8
Cross-cultural entity translation remains challenging for large language models (LLMs) as literal or phonetic renderings are usually yielded instead of culturally appropriate trans…
Uncovering the Fragility of Trustworthy LLMs through Chinese Textual Ambiguity
Xinwei Wu, Haojie Li, Hongyu Liu +4
In this work, we study a critical research problem regarding the trustworthiness of large language models (LLMs): how LLMs behave when encountering ambiguous narrative text, with a…
Challenging Multilingual LLMs: A New Taxonomy and Benchmark for Unraveling Hallucination in Translation
Xinwei Wu, Heng Liu, Jiang Zhou +5
Large Language Models (LLMs) have advanced machine translation but remain vulnerable to hallucinations. Unfortunately, existing MT benchmarks are not capable of exposing failures i…
: Multi-Perspective Multi-Pair Preference Optimization for Machine Translation
Hao Wang, Linlong Xu, Heng Liu +12
Aligning Large Language Models (LLMs) with human preferences is pivotal for Machine Translation (MT), yet current approaches are often hindered by misleading reward signals. Our an…