2026
CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation
Yutong Song, Jiang Wu, Weijia Zhang, Chengze Shen, Shaofan Yuan, Weitao Lu, Jian Wang, Yu Wang, Nikil Dutt, Amir M. Rahmani
arXiv preprint 2026
A hierarchical personalization framework that clusters users by stylistic patterns, learns group-specific LoRA adapters, and injects individual preferences only at decoding time while keeping the base model frozen.
CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation
Yutong Song, Jiang Wu, Weijia Zhang, Chengze Shen, Shaofan Yuan, Weitao Lu, Jian Wang, Yu Wang, Nikil Dutt, Amir M. Rahmani
arXiv preprint 2026
A hierarchical personalization framework that clusters users by stylistic patterns, learns group-specific LoRA adapters, and injects individual preferences only at decoding time while keeping the base model frozen.
2025
Summarize-Exemplify-Reflect: Data-driven Insight Distillation Empowers LLMs for Few-shot Tabular Classification
Yifei Yuan, Jiatong Li, Weijia Zhang, Mohammad Aliannejadi, Evangelos Kanoulas, Renjun Hu
Findings of EMNLP 2025 2025 Findings
InsightTab distills tabular data into actionable insights via rule summarization, strategic exemplification, and reflection, consistently improving LLM few-shot tabular classification across nine datasets.
Summarize-Exemplify-Reflect: Data-driven Insight Distillation Empowers LLMs for Few-shot Tabular Classification
Yifei Yuan, Jiatong Li, Weijia Zhang, Mohammad Aliannejadi, Evangelos Kanoulas, Renjun Hu
Findings of EMNLP 2025 2025 Findings
InsightTab distills tabular data into actionable insights via rule summarization, strategic exemplification, and reflection, consistently improving LLM few-shot tabular classification across nine datasets.
Beyond Natural Language Plans: Structure-Aware Planning for Query-Focused Table Summarization
Weijia Zhang, Songgaojun Deng, Evangelos Kanoulas
arXiv preprint 2025
Replaces ambiguous natural-language plans with TaSoF, a structured plan executed as a dependency graph of SQL steps, making multi-table query-focused summarization more reliable and scalable.
Beyond Natural Language Plans: Structure-Aware Planning for Query-Focused Table Summarization
Weijia Zhang, Songgaojun Deng, Evangelos Kanoulas
arXiv preprint 2025
Replaces ambiguous natural-language plans with TaSoF, a structured plan executed as a dependency graph of SQL steps, making multi-table query-focused summarization more reliable and scalable.
2024
Beyond Relevant Documents: A Knowledge-Intensive Approach for Query-Focused Summarization using Large Language Models
Weijia Zhang, Jia-Hong Huang, Svitlana Vakulenko, Yumo Xu, Thilina Rajapakse, Evangelos Kanoulas
International Conference on Pattern Recognition (ICPR) 2024
Reframes query-focused summarization as a knowledge-intensive task: a retrieval module pulls candidate documents from a large corpus, removing the usual assumption that relevant documents are given in advance.
Beyond Relevant Documents: A Knowledge-Intensive Approach for Query-Focused Summarization using Large Language Models
Weijia Zhang, Jia-Hong Huang, Svitlana Vakulenko, Yumo Xu, Thilina Rajapakse, Evangelos Kanoulas
International Conference on Pattern Recognition (ICPR) 2024
Reframes query-focused summarization as a knowledge-intensive task: a retrieval module pulls candidate documents from a large corpus, removing the usual assumption that relevant documents are given in advance.
QFMTS: Generating Query-Focused Summaries over Multi-Table Inputs
Weijia Zhang, Vaishali Pal, Jia-Hong Huang, Evangelos Kanoulas, Maarten de Rijke
European Conference on Artificial Intelligence (ECAI) 2024
Introduces query-focused multi-table summarization, along with a dataset of 4,909 query-summary pairs and a table serialization plus summarization-controller pipeline that tailors summaries to a user's query.
QFMTS: Generating Query-Focused Summaries over Multi-Table Inputs
Weijia Zhang, Vaishali Pal, Jia-Hong Huang, Evangelos Kanoulas, Maarten de Rijke
European Conference on Artificial Intelligence (ECAI) 2024
Introduces query-focused multi-table summarization, along with a dataset of 4,909 query-summary pairs and a table serialization plus summarization-controller pipeline that tailors summaries to a user's query.
Towards Fine-Grained Citation Evaluation in Generated Text: A Comparative Analysis of Faithfulness Metrics
Weijia Zhang, Mohammad Aliannejadi, Yifei Yuan, Jiahuan Pei, Jia-Hong Huang, Evangelos Kanoulas
International Natural Language Generation Conference (INLG) 2024 Oral
A comparative evaluation framework showing how well faithfulness metrics separate full, partial, and no citation support, moving citation evaluation past binary classification.
Towards Fine-Grained Citation Evaluation in Generated Text: A Comparative Analysis of Faithfulness Metrics
Weijia Zhang, Mohammad Aliannejadi, Yifei Yuan, Jiahuan Pei, Jia-Hong Huang, Evangelos Kanoulas
International Natural Language Generation Conference (INLG) 2024 Oral
A comparative evaluation framework showing how well faithfulness metrics separate full, partial, and no citation support, moving citation evaluation past binary classification.
FoodieQA: A Multimodal Dataset for Fine-Grained Understanding of Chinese Food Culture
Wenyan Li, Xinyu Zhang, Jiaang Li, Qiwei Peng, Raphael Tang, Li Zhou, Weijia Zhang, et al.
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2024
A manually curated image-text dataset covering regional Chinese food culture, on which open-weights VLMs still trail humans by 41% on multi-image and 21% on single-image VQA.
FoodieQA: A Multimodal Dataset for Fine-Grained Understanding of Chinese Food Culture
Wenyan Li, Xinyu Zhang, Jiaang Li, Qiwei Peng, Raphael Tang, Li Zhou, Weijia Zhang, et al.
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2024
A manually curated image-text dataset covering regional Chinese food culture, on which open-weights VLMs still trail humans by 41% on multi-image and 21% on single-image VQA.
A Comparative Analysis of Faithfulness Metrics and Humans in Citation Evaluation
Weijia Zhang, Mohammad Aliannejadi, Jiahuan Pei, Yifei Yuan, Jia-Hong Huang, Evangelos Kanoulas
LLM4Eval @ SIGIR — The First Workshop on Large Language Model for Evaluation in Information Retrieval 2024
Workshop version of our fine-grained citation evaluation study, comparing faithfulness metrics against human judgements across three levels of citation support.
A Comparative Analysis of Faithfulness Metrics and Humans in Citation Evaluation
Weijia Zhang, Mohammad Aliannejadi, Jiahuan Pei, Yifei Yuan, Jia-Hong Huang, Evangelos Kanoulas
LLM4Eval @ SIGIR — The First Workshop on Large Language Model for Evaluation in Information Retrieval 2024
Workshop version of our fine-grained citation evaluation study, comparing faithfulness metrics against human judgements across three levels of citation support.
2023
Tackling Query-Focused Summarization as A Knowledge-Intensive Task: A Pilot Study
Weijia Zhang, Svitlana Vakulenko, Thilina Rajapakse, Yumo Xu, Evangelos Kanoulas
Gen-IR @ SIGIR — The First Workshop on Generative Information Retrieval 2023
Builds KI-QFS, a benchmark where answering a query requires retrieving from a knowledge corpus first, and benchmarks state-of-the-art QFS and retrieval-enhanced models on it.
Tackling Query-Focused Summarization as A Knowledge-Intensive Task: A Pilot Study
Weijia Zhang, Svitlana Vakulenko, Thilina Rajapakse, Yumo Xu, Evangelos Kanoulas
Gen-IR @ SIGIR — The First Workshop on Generative Information Retrieval 2023
Builds KI-QFS, a benchmark where answering a query requires retrieving from a knowledge corpus first, and benchmarks state-of-the-art QFS and retrieval-enhanced models on it.
2019
A Deep Neural Information Fusion Architecture for Textual Network Embeddings
Zenan Xu, Qinliang Su, Xiaojun Quan, Weijia Zhang
Conference on Empirical Methods in Natural Language Processing (EMNLP-IJCNLP) 2019
A deep architecture that fuses structural and textual signals into a single network embedding through a new objective, a complementary fusion method, and a mutual gate mechanism.
A Deep Neural Information Fusion Architecture for Textual Network Embeddings
Zenan Xu, Qinliang Su, Xiaojun Quan, Weijia Zhang
Conference on Empirical Methods in Natural Language Processing (EMNLP-IJCNLP) 2019
A deep architecture that fuses structural and textual signals into a single network embedding through a new objective, a complementary fusion method, and a mutual gate mechanism.