WisPaper: Your AI Scholar Search Engine

arXiv:2512.06879v3 Announce Type: replace-cross Abstract: We present extsc{WisPaper}, an end-to-end agent system that transforms how researchers discover, organize, and track academic literature. The system addresses two fundamental challenges. (1)~extit{Semantic search limitations}: existing academic search engines match keywords but cannot verify whether papers truly address complex research questions; and (2)~extit{Workflow fragmentation}: researchers must manually stitch together separate tools for discovery, organization, and monitoring. extsc{WisPaper} tackles these through three integrated modules. extbf{Scholar Search} combines rapid keyword retrieval with extit{Deep Search}, in which an agentic model, extsc{WisModel}, validates candidate papers against user queries through structured reasoning. Discovered papers flow seamlessly into extbf{Library} with one click, where systematic organization progressively builds a user profile that sharpens the recommendations of extbf{AI Feeds}, which continuously surfaces relevant new publications and in turn guides subsequent exploration, closing the loop from discovery to long-term awareness. On TaxoBench, extsc{WisPaper} achieves 22.26% recall, surpassing the O3 baseline (20.92%). Furthermore, extsc{WisModel} attains 93.70% validation accuracy, effectively mitigating retrieval hallucinations.

Related

• Beyond RAG for Agent Memory: Retrieval by Decoupling and Aggregation
• El Agente Quntur: A research collaborator agent for quantum chemistry
• PosterGen: Aesthetic-Aware Multi-Modal Paper-to-Poster Generation via Multi-Agent LLMs
• Process-Centric Analysis of Agentic Software Systems

Source: arXiv cs.AI | 2026-04-14