Pasa

PaSa: An LLM Agent for Comprehensive Academic Paper Search
ACL 2025 (Main)

ByteDance Seed

[](https://arxiv.org/abs/2501.10120) [](https://pasa-agent.ai) [](https://huggingface.co/bytedance-research/pasa-7b-crawler) [](https://huggingface.co/bytedance-research/pasa-7b-selector) [](https://huggingface.co/datasets/CarlanLark/pasa-dataset)

Introduction

We introduce PaSa, an advanced **Pa**per **S**e**a**rch agent powered by large language models. PaSa can autonomously make a series of decisions, including invoking search tools, reading papers, and selecting relevant references, to ultimately obtain comprehensive and accurate results for complex scholarly queries. We optimize PaSa using reinforcement learning with a synthetic dataset, AutoScholarQuery, which includes 35k fine-grained academic queries and corresponding papers sourced from top-tier AI conference publications. Additionally, we develop RealScholarQuery, a benchmark collecting real-world academic queries to assess PaSa performance in more realistic scenarios. Despite being trained on synthetic data, PaSa significantly outperforms existing baselines on RealScholarQuery, including Google, Google Scholar, Google with GPT-4 for paraphrased queries, chatGPT (search-enabled GPT-4o), GPT-o1, and PaSa-GPT-4o (PaSa implemented by prompting GPT-4o). Notably, PaSa-7B surpasses the best Google-based baseline, Google with GPT-4o, by 37.78% in recall@20 and 39.90% in recall@50. It also exceeds PaSa-GPT-4o by 30.36% in recall and 4.25% in precision.

Qui