WisPaper: Your AI Scholar Search Engine

We present $\text{textsc}WisPaper$, an end-to-end agent system that transforms how researchers discover, organize, and track academic literature. The system addresses two fundamental challenges. (1)~$\mathit{\text{Semantic}}\mathit{\text{search}}\mathit{\text{limitations}}$: existing academic search engines match keywords but cannot verify whether papers truly address complex research questions; and (2)~$\mathit{\text{Workflow}}\mathit{\text{fragmentation}}$: researchers must manually stitch together separate tools for discovery, organization, and monitoring. $\text{textsc}WisPaper$ tackles these through three integrated modules. $\mathbf{\text{Scholar}}\mathbf{\text{Search}}$ combines rapid keyword retrieval with $\mathit{\text{Deep}}\mathit{\text{Search}}$, in which an agentic model, $\text{textsc}WisModel$, validates candidate papers against user queries through structured reasoning. Discovered papers flow seamlessly into $\mathbf{\text{Library}}$ with one click, where systematic organization progressively builds a user profile that sharpens the recommendations of $\mathbf{\text{AI}}\mathbf{\text{Feeds}}$, which continuously surfaces relevant new publications and in turn guides subsequent exploration, closing the loop from discovery to long-term awareness. On TaxoBench, $\text{textsc}WisPaper$ achieves 22.26% recall, surpassing the O3 baseline (20.92%). Furthermore, $\text{textsc}WisModel$ attains 93.70% validation accuracy, effectively mitigating retrieval hallucinations.