Information Retrieval

Yuanhong Zheng, Ruichuan An, Xiaopeng Lin et al.

This addresses the limitation of current personalization methods to static/offline data for future AI assistants, though it is incremental as it builds on existing vision-language models.

MiroMind Team, S. Bai, L. Bing et al.

This addresses the problem of reliable multi-step problem solving for research agents, though it appears incremental as it builds on prior agent frameworks.

Negar Arabzadeh, Wenjie Ma, Sewon Min et al.

For researchers and practitioners in reasoning-intensive tasks like math and code generation, this work provides a simple yet effective method to improve performance via RAG with thinking traces, challenging a widely held belief.

Zhuofeng Li, Dongfu Jiang, Xueguang Ma et al.

This addresses reproducibility and cost issues for researchers developing deep research agents, though it is incremental as it builds on existing methods with a novel pipeline design.

Minghao Guo, Qingyue Jiao, Zeru Shi et al.

For researchers developing long-term multimodal memory systems, MemEye provides a rigorous benchmark to identify critical bottlenecks in visual evidence preservation and temporal reasoning.

Nandan Thakur, Zijian Chen, Xueguang Ma et al.

This addresses the problem of expensive data annotation for search agents, though it is incremental as it builds on existing synthetic data generation methods.

Xuanwang Zhang, Yuteng Han, Jinnan Qi et al.

This work solves the problem of inefficient web navigation for AI agents by overcoming a key bottleneck, with significant performance gains in complex environments.

Xiaoyan Zhao, Juntao You, Yang Zhang et al.

This addresses the challenge of effectively personalizing LLMs for individual users in real-world applications, representing a novel method for a known bottleneck.

Shubham Kumar Nigam, Suparnojit Sarkar, Piyush Patel

For healthcare AI researchers and practitioners in India, this work provides a realistic multilingual medical dialogue resource and a fine-tuned model, though the approach is incremental (extending existing datasets and fine-tuning a small model).

OneRec Team, Biao Yang, Boyang Ding et al.

This work tackles the problem of enabling chain-of-thought reasoning in generative recommendation for practitioners deploying such models in large-scale platforms.

Yingyi Zhang, Junyi Li, Wenlin Zhang et al.

This work addresses the challenge of scalable and effective personalization for LLM users, offering an incremental improvement over existing retrieval methods.

Zhichao Xu, Fengran Mo, Zhiqi Huang et al.

This is an incremental survey paper that synthesizes existing research on model architectures in information retrieval for researchers and practitioners in the field.

Jinheon Baek, Soyeong Jeong, Sangwoo Park et al.

For practitioners needing to query diverse knowledge sources, OmniRetrieval provides a general-purpose interface that preserves structural affordances without homogenization.

Changxin Lao, Fei Pan, Guozhuang Ma et al.

For industrial recommender systems, AgentX addresses the bottleneck of human-dependent iteration, enabling autonomous and scalable experimentation.

Zhuofeng Li, Haoxiang Zhang, Cong Wei et al.

For agentic search tasks, where conventional retrieval interfaces limit multi-step reasoning, DCI provides a more flexible interface that improves retrieval quality as language agents become stronger.

Yi Yu, Junzhuo Ma, Chenghuang Shen et al.

This work provides a practical solution for deploying efficient technical service agents, though it appears incremental as it builds on existing adaptation methods.

Buqiang Xu, Yijun Chen, Jizhan Fang et al.

For developers of long-term conversational agents, StructMem offers a more efficient memory system that balances relational structure with computational cost.

Chenxi Wang, Zhuoyun Yu, Xin Xie et al.

This addresses the problem of redundant exploration and poor generalization in LLM agents for AI researchers, offering a plug-and-play solution to enhance agent learning.

Yijie Ding, Zitian Guo, Jiacheng Li et al.

This work addresses the problem of understanding and improving recommendation system generalization for researchers and practitioners, though it is incremental in nature.

Duyi Pan, Tianao Lou, Xin Li et al.

This addresses recall and precision limitations in graph-based RAG for black-box knowledge graphs, offering a plug-and-play solution for knowledge-intensive tasks.