🧑‍🔬 Ji-Rong Wen

Affiliations

• Renmin University of China · Beijing, China — current · 2013–2026 on 517 papers · employment 2013–2016: Professor, School of Information (ORCID) · employment from 2016: Professor, Dean, School of Information (ORCID) · employment from 2019: Professor, Executive Dean, Gaoling School of Artificial Intelligence (ORCID)
• Ministry of Education of the People's Republic of China · Beijing, China — 2021–2024 on 9 papers
• Beijing Academy of Artificial Intelligence · Beijing, China — 2021–2022 on 4 papers
• Chinese Academy of Sciences · Beijing, China — 2000–2022 on 2 papers
• Association for Computing Machinery · New York, United States — 2019 on 2 papers
• Microsoft Research Asia (China) · Beijing, China — 1999–2017 on 94 papers · employment 1999–2013: Associate Researcher, Researcher, Senior Researcher/Research Manager (ORCID)
• Microsoft (United States) · Redmond, United States — 2000–2010 on 5 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 13 more institutions on single bylines are not listed.

Topics

Language models & pretraining6Evaluation & benchmarks2Information retrieval & search2Discourse analysis1Dialogue systems1Multilingual & cross-lingual NLP1Vision & language1

Papers

The most cited first (OpenAlex and our library).

1. Evaluating Object Hallucination in Large Vision-Language Models (2023 · 986 citations)
   Large vision-language models (LVLMs), which connect a powerful LLM to a visual encoder, often describe objects that are not actually in the image.
2. HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models (2023 · 646 citations)
   Large language models such as ChatGPT often produce content that contradicts its source or cannot be checked against known facts.
3. Large Language Models for Information Retrieval: A Survey (2023 · 335 citations)
   This survey maps how large language models are changing text information retrieval. The authors start from a known tension. Classic term-based methods such as BM25 are fast. Neural models capture meaning better but suffer from scarce training data, weak interpretability and plausible-sounding errors. LLMs like…
4. RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking (2021 · 256 citations)
   Retrieve-then-rerank pipelines pair a fast dual-encoder retriever with a stronger cross-encoder re-ranker.
5. Large Language Diffusion Models (2025 · 164 citations)
   Modern LLMs almost all rely on autoregressive next-token prediction. Many assume that scaling, in-context learning and instruction-following depend on this design. The authors test that assumption with LLaDA, a masked diffusion language model trained from scratch.
6. Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models (2024 · 151 citations)
   Large language models handle many languages well even though they are trained mostly on English and without curated parallel data, and it is unclear how they do this internally.
7. The Dawn After the Dark: An Empirical Study on Factuality Hallucination in Large Language Models (2024 · 150 citations)
   Large language models often produce plausible-sounding but factually wrong content, which limits their use in sensitive settings.
8. R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning (2025 · 137 citations)
   Reasoning-oriented LLMs trained with RL perform well on maths and code, but they depend on what they memorised.
9. A Joint Model for Dropped Pronoun Recovery and Conversational Discourse Parsing in Chinese Conversational Speech (2021 · 16 citations)
   Chinese speakers frequently omit pronouns that listeners can infer from context. Existing systems handle two related problems separately: dropped pronoun recovery (DPR), which finds where a pronoun is missing and what type it is, and conversational discourse parsing (CDP), which links utterances through relations…

Also in our library

19 more works of theirs are in our library without a page here yet; they will appear as we publish them.

Co-authors here

12 of 23 co-authors with a page, the most papers together first.

• Wayne Xin Zhao ×3
• Xin Zhao ×3
• Jian-Yun Nie ×2
• Jie Chen ×2
• Junyi Li ×2
• Ruiyang Ren ×2
• Xiaoxue Cheng ×2
• Dongdong Zhang
• Furu Wei
• Haifeng Wang
• Haoyang Huang
• Hua Wu

Cited by

Papers of our library cite their works 3 847 times (of the 140 884 papers whose bibliographies we read); 172 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/ji-rong-wen/