Xu Jinan
Xu Jinan has 4 papers in our library, in 2026, most of them on Multilingual & cross-lingual NLP, Language models & pretraining, Evaluation & benchmarks; often with Kaiyu Huang, Yufeng Chen, Fengran Mo.
OpenAlex · h-index 0 · 4 works · 0 citations OpenAlex, 2026-10-14
📄 4 works in our library · 🔗 Cited 17 times in our library · ⭐ Paper Stars ranking: #10481
Affiliations
Fields, topics and concepts
📚 Shelves 8The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 4 worksComputer Science · 4 worksArtificial Intelligence · 4 worksLanguage and Linguistics · 4 worksComputer Vision and Pattern Recognition · 2 worksInformation Systems · 1 workSocial Sciences · 1 workSafety Research · 1 work
🧭 Topics 7Research subjects and languages the work is about.Multilingual & cross-lingual NLP · 2 worksReasoning · 1 workInformation retrieval & search · 1 workBias, fairness & safety · 1 workEvaluation & benchmarks · 1 workLanguage models & pretraining · 1 workVision & language · 1 work
Works by year
2020s
Papers
- Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning (2026 · 11 citations)
Reasoning-oriented LLMs that think before answering often handle non-English prompts poorly. They tend to drift into English or a mix of languages during reasoning, and their accuracy drops compared with English. The authors introduce M-Thinker, a family of models (1.5B, 4B, 7B) trained with GRPO using two new reward… - Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation (2026 · 5 citations)
The paper addresses two problems in multilingual retrieval-augmented generation (MRAG). The first is knowledge bias: an LLM answers the same question differently depending on its language. The second is knowledge conflict: passages retrieved from collections in many languages contradict each other. The authors… - When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily Life (2026 · 1 citation)
Multimodal LLMs are increasingly used as everyday assistants. Their answers can steer people into risky actions even when the user has no harmful intent. Most existing safety benchmarks target jailbreaks or explicit harm, and implicit-risk sets tend to be small or artificial. The authors introduce SaLAD, a benchmark… - Can Multi-agent Help Disambiguation in Multi-domain Translation? (2026)
The paper asks whether LLM-based multi-agent collaboration can resolve cross-domain word ambiguity in multi-domain translation (MDT).
Co-authors here
- Kaiyu Huang ×3
- Yufeng Chen ×3
- Fengran Mo ×2
- Xue Zhang ×2
- Fandong Meng
- Hongliang Li
- Jian-Yun Nie
- Jie Zhou
- Meng Jiang
- Rui Qi
- Shaoyang Xu
- Shuo Wang
Venues
- Findings of the Association for Computational Linguistics ×3: Findings of the Association for Computational Linguistics 2026 ×3
- Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2026
- Workshop on Multilinguality in the Era of Large Language Models
Cited by
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/authors/xu-jinan/