News

(Aug 2026) ⭐Our survey paper for Context Compression for LLM Agents "Context Compression for LLM Agents: A Survey of Methods, Failure Modes, and Evaluation" is accepted to EMNLP 2026 Main, quote EMNLP email: "This year, we received an unprecedented 17669 submissions. We accepted 2719 as Main Conference papers and 2533 to Findings of the ACL. This represents an acceptance rate of 15.4% for Main Conference papers and 14.3% for Findings papers."

(Aug 2026) ⭐Check out our survey paper for Context Compression for LLM Agents! "Context Compression for LLM Agents: A Survey of Methods, Failure Modes, and Evaluation" (Let us know your related papers and any suggestions) [Link: https://github.com/YerbaPage/Awesome-Agent-Context-Compression]

(Aug 2026) ⭐"LQ-Rec: Late Quantization for Sequential Recommendation" is accepted to ICDM 2026, see everyone in Shenyang, China on Nov 12

(July 2026) ⭐WebReal is accepted at COLM 2026 for benchmarking DeepSearch Agents with real-world user information.

(July 2026) ⭐ACL 2026 will be held in San Diego, wish to see all friends here.🌊

(June 2026)⭐We have a paper accepted to ECCV 2026 about fine-grained video understanding and attention sink!

(Jan 2026) ⭐I will join Meta AI as a research intern starting summer 2026, see everyone at Silicon Valley!

(Jan 2026) ⭐We have a paper accepted to ICLR 2026 on Frequency-Aware-Sparse-Attention (FASA), revealing key attributes of RoPE-based positional encoding that a small dominant set of frequency dimensions exhibits high overlap with full attention for token selection. +2.56x acceleration is achieved compared to full attention.

(Dec 2025) ❀ NeurIPS 2025 will be held in San Diego, see you here!🌊

(May 2025) ❀ We have a paper accepted to KDD 2025 (Oral) on Preference-Optimized Retrieval and Ranking for Efficient Multimodal Recommendation (PRIME). 

(Sep 2025) ❀ We have a paper accepted to EMNLP 2025 Main on Multimodal Retrieval Augmented Generation (RAG)-based Direct Preference Optimization (rDPO) with enhanced vision-language alignment and improved visual question answering (VQA).

(Nov, 2024) ❀ Our work of Auto-Encoding vs. Auto-Regression for self-attentive sequential recommendation as a reality check, Your Causal Self-Attentive Recommender Hosts a Lonely Neighborhood, is accepted to WSDM 2025 (Oral), see everyone in Germany, March 2025!

(Oct, 2024) ❀ Our work of fMRLRec, Train Once, Deploy Anywhere: Matryoshka Representation Learning for Multimodal Recommendation, is accepted to Findings of EMNLP 2024, see everyone in Miami, November 2024!

(2024) ❀ Our work of LRURec, Linear Recurrent Units for Sequential Recommendation, is accepted to WSDM 2024, see everyone in Mexico, March 2024!

Publications (To be updated)

Your Causal Self-Attentive Recommender Hosts a Lonely Neighborhood. (2025) [Oral]

Yueqi Wang, Zhankui He, Zhenrui Yue, Julian McAuley, Dong Wang

Proceedings of the 18th ACM International Conference on Web Search and Data Mining (WSDM).

[Code, PDF]

Train Once, Deploy Anywhere: Matryoshka Representation Learning for Multimodal Recommendation. (2024)

Yueqi Wang, Zhenrui Yue, Huimin Zeng, Dong Wang, Julian McAuley

Findings of the Association for Computational Linguistics: EMNLP 2024 (EMNLP Findings).

[Code, PDF]

Linear Recurrent Units for Sequential Recommendation. (2024)

Zhenrui Yue, Yueqi Wang, Zhankui He, Huimin Zeng, Julian McAuley, Dong Wang.

Proceedings of the 17th ACM International Conference on Web Search and Data Mining (WSDM).

[Code, PDF, Blog]

Preference-Optimized Retrieval and Ranking for Efficient Multimodal Recommendation. (2025) [Oral]

Zhenrui Yue, Huimin Zeng, Yueqi Wang, Julian McAuley, Dong Wang.

Proceedings of the 31th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD).

Learning from Mistakes: Contrastive Decoding for Retrieval Augmented Generation. (Under review)

Zhenrui Yue, Huimin Zeng, Yueqi Wang, Yaokun Liu, Fengran Mo, Yang Zhang, Na Wei, Dong Wang

Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL).

Multimodal Cross-Correlation Learning for Fair Vision-Language Models in Federated Learning. (Under review)

Huimin Zeng, Zhenrui Yue, Yueqi Wang, Yang Zhang, Dong Wang

Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL).

Does Chronology Matter? Sequential vs Contextual Approaches to Knowledge Tracing. (2022)

Yueqi Wang, Zachary Pardos

Proceedings of the 15th International Conference on Educational Data Mining (EDM), 601–605.

[PDF]

BertKT: A Purely Attention-Based, Bidirectional Deep Learning Architecture for Knowledge Tracing. (2020)

Yueqi Wang, Zachary Pardos

Proceedings of the Workshop on AI-based Multimodal Analytics for Understanding Human Learning in Real-world Educational Contexts (AIMA4EDU) @ the 29th International Joint Conference on Artificial Intelligence (IJCAI).

Experiences

University of California, San Diego | San Diego, CA | Aug 2023 - Now | (Research Intern)

• With Professor Julian McAuley.

University of California, Berkeley | Berkeley, CA | Aug 2022 - Dec 2022 | Paid position

• Lead Graduate Student Instructor (Lead GSI) for DATA 144/EDUC 244 Data Mining and Analytics.

University of California, Berkeley | Berkeley, CA | Jan 2022 - Aug 2022 | Paid position

• Graduate Student Researcher (GSR) with Prof. Zachary Pardos, Knowledge Tracing project.

University of electronic Science and Technology of China (UESTC) | Chengdu, China | July 2020 - June 2021 | Paid position

• Research Assistant