Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

WisPaper: Your AI Scholar Search Engine

Li Ju, Jun Zhao, Mingxu Chai, Ziyu Shen, Xiangyang Wang, Yage Geng, Chunchun Ma, Hao Peng, Guangbin Li, Tao Li, Chengyong Liao, Fu Wang, Xiaolong Wang, Junshen Chen, Rui Gong, Shijia Liang, Feiyan Li, Ming Zhang, Kexin Tan, Jujie Ye, Zhiheng Xi, Shihan Dou, Tao Gui, Yuankai Ying, Yang Shi, Yue Zhang, Qi Zhang

TL;DR

WisPaper addresses the challenge of exponential literature growth by delivering an integrated platform that unifies discovery, organization, and frontier tracking. It combines dual-mode Scholar Search (quick keyword-based retrieval and deep agent-powered reasoning) with a configurable Library and an AI Feeds recommendation system, forming a closed-loop workflow. The WisModel training pipeline pairs supervised fine-tuning with Group Relative Policy Optimization to enable reliable, criteria-driven paper validation. Evaluations on 2,777 queries and 5,879 human-annotated criteria show state-of-the-art performance in both query interpretation and paper-criteria matching, surpassing strong baselines and demonstrating substantial reductions in screening and management effort. Overall, WisPaper enables multilingual, multidisciplinary researchers to conduct end-to-end literature workflows more efficiently and systematically.

Abstract

Researchers struggle to efficiently locate and manage relevant literature within the exponentially growing body of scientific publications. We present \textsc{WisPaper}, an intelligent academic retrieval and literature management platform that addresses this challenge through three integrated capabilities: (1) \textit{Scholar Search}, featuring both quick keyword-based and deep agentic search modes for efficient paper discovery; (2) \textit{Library}, a customizable knowledge base for systematic literature organization; and (3) \textit{AI Feeds}, an intelligent recommendation system that automatically delivers relevant new publications based on user interests. Unlike existing academic tools, \textsc{WisPaper} provides a closed-loop workflow that seamlessly connects literature discovery, management, and continuous tracking of research frontiers. Our multilingual and multidisciplinary system significantly reduces the time researchers from diverse backgrounds spend on paper screening and management, enabling them to focus on their core research activities. The platform is publicly accessible and serves researchers across academia and industry.

WisPaper: Your AI Scholar Search Engine

TL;DR

WisPaper addresses the challenge of exponential literature growth by delivering an integrated platform that unifies discovery, organization, and frontier tracking. It combines dual-mode Scholar Search (quick keyword-based retrieval and deep agent-powered reasoning) with a configurable Library and an AI Feeds recommendation system, forming a closed-loop workflow. The WisModel training pipeline pairs supervised fine-tuning with Group Relative Policy Optimization to enable reliable, criteria-driven paper validation. Evaluations on 2,777 queries and 5,879 human-annotated criteria show state-of-the-art performance in both query interpretation and paper-criteria matching, surpassing strong baselines and demonstrating substantial reductions in screening and management effort. Overall, WisPaper enables multilingual, multidisciplinary researchers to conduct end-to-end literature workflows more efficiently and systematically.

Abstract

Researchers struggle to efficiently locate and manage relevant literature within the exponentially growing body of scientific publications. We present \textsc{WisPaper}, an intelligent academic retrieval and literature management platform that addresses this challenge through three integrated capabilities: (1) \textit{Scholar Search}, featuring both quick keyword-based and deep agentic search modes for efficient paper discovery; (2) \textit{Library}, a customizable knowledge base for systematic literature organization; and (3) \textit{AI Feeds}, an intelligent recommendation system that automatically delivers relevant new publications based on user interests. Unlike existing academic tools, \textsc{WisPaper} provides a closed-loop workflow that seamlessly connects literature discovery, management, and continuous tracking of research frontiers. Our multilingual and multidisciplinary system significantly reduces the time researchers from diverse backgrounds spend on paper screening and management, enabling them to focus on their core research activities. The platform is publicly accessible and serves researchers across academia and industry.

Paper Structure

This paper contains 23 sections, 2 equations, 2 figures, 10 tables.

Table of Contents

  1. Introduction
  2. WisPaper
  3. Scholar Search Module
  4. Quick Search: Millisecond-Level Responsive Literature Retrieval
  5. Deep Search: Agent-Powered Intelligent Literature Exploration
  6. Library Module
  7. AI Feeds Module
  8. Module Interconnection
  9. Deep Search: WisModel Training Pipeline
  10. Training Objectives
  11. Objective 1: Query Understanding and Criteria Generation.
  12. Objective 2: Paper-Criteria Matching Assessment.
  13. Training Procedure
  14. Stage 1: Supervised Fine-Tuning.
  15. Stage 2: Group Relative Policy Optimization.
  16. ...and 8 more sections

Figures (2)

  • Figure 1: WisPaper Knowledge Loop
  • Figure 2: Distribution statistics of the evaluation dataset across 10 academic disciplines. The left panel shows the average query length in tokens for each discipline. The right panel displays the average number of criteria per query. These statistics indicate the complexity of search requirements across different fields.