Information Retrieval

Weizhi Zhang, Yangning Li, Yuanchen Bei et al. · pku

This addresses the problem of handling complex, multi-step information retrieval for users across diverse domains, representing a potential paradigm shift rather than an incremental improvement.

Weihang Su, Yichen Tang, Qingyao Ai et al.

This addresses the problem of improving reliability and performance in large language models for applications requiring up-to-date or domain-specific knowledge, representing a novel paradigm rather than an incremental improvement.

Juntao Tan, Shuyuan Xu, Wenyue Hua et al.

This work addresses the problem of aligning LLMs with recommendation needs for researchers and practitioners in AI and recommender systems, offering a novel paradigm that is incremental in advancing generative recommendation techniques.

Teo Susnjak, Peter Hwang, Napoleon H. Reyes et al.

This foundational study addresses the labor-intensive process of literature reviews for academic researchers, advocating for updated guidelines to incorporate AI-driven methods.

Geonmo Gu, Sanghyuk Chun, Wonjae Kim et al.

This addresses the data efficiency problem for researchers and practitioners in image retrieval by enabling zero-shot composed retrieval without expensive triplet data.

Yan Weng, Fengbin Zhu, Tong Ye et al.

This work addresses the problem of effective knowledge integration in Large Language Models for more accurate and reliable responses, which is significant for natural language processing applications.

Orion Weller, Kathryn Ricci, Eugene Yang et al.

This addresses the need for more efficient and explainable reranking models in search systems, representing a novel approach rather than an incremental improvement.

Yundong Sun, Dongjie Zhu, Yansong Wang et al.

This work addresses a key bottleneck in graph representation learning for researchers and practitioners, offering a novel collaborative approach to improve model depth and accuracy.

Shanshan Zhong, Zhongzhan Huang, Daifeng Li et al.

This addresses robustness issues for users of multimodal recommender systems, offering a novel paradigm that complements existing methods.

Sam Lin, Wenyue Hua, Zhenting Wang et al.

It addresses privacy concerns for users of cloud-based LLMs like ChatGPT, offering a practical solution to prevent data exposure to service providers and jailbreaking attacks.

Guoxuan Chen, Lianghao Xia, Chao Huang

This addresses the need for efficient data removal in recommender systems due to privacy and regulatory requirements, offering a practical solution for incremental improvements in unlearning efficiency.

Haoyu Wang, Sunhao Dai, Haiyuan Zhao et al.

This addresses source bias in information retrieval systems, which threatens the information access ecosystem, by providing a novel causal explanation and solution.

Abdelrahman Abdallah, Mohammed Ali, Bhawna Piryani et al.

This addresses the problem of deep semantic inference in retrieval for users needing complex, multi-step queries, representing a novel paradigm rather than an incremental improvement.

Fabian Paischer, Liu Yang, Linfeng Liu et al.

This addresses the issue of echo chambers and inflexibility in recommendation systems for users, though it is incremental as it builds on existing generative models.

Zirui Tang, Boyu Niu, Xuanhe Zhou et al.

This addresses the costly and inefficient reliance on human analysts for interpreting semi-structured tables in domains like finance and healthcare, offering a novel automated solution.

Tiansheng Wen, Yifei Wang, Zequn Zeng et al.

This work addresses the need for efficient and high-fidelity adaptive embeddings in real-world applications like retrieval and search, offering a novel alternative to existing methods.

Zhengyang Su, Isay Katsman, Yueqi Wang et al.

This work addresses the problem of inefficient constrained decoding for LLM-based generative retrieval on hardware accelerators, which is critical for industrial recommender systems requiring output space restrictions.

Amin Bigdeli, Negar Arabzadeh, Ebrahim Bagheri et al.

This addresses a security threat for neural retrieval systems by demonstrating a scalable attack method that works without gradient access.

Chao-Wei Huang, Chen-An Li, Tsu-Yuan Hsu et al.

This addresses the challenge of data scarcity in multilingual information retrieval, making dense retrievers more practical for applications where paired data is unavailable.

Bangrui Xu, Qihang Yao, Zirui Tang et al.

This work is significant for researchers and practitioners working with semi-structured documents, as it offers a substantial improvement in accuracy for natural language question answering, addressing key limitations of existing methods.