Academia AI Integration

Search arXiv papers with field-specific queries. Fields: all: (all fields), ti: (title), au: (author), abs: (abstract), cat: (category), id: (ID without version) Operators: AND, OR, ANDNOT Please always specify the fields. Search should always be field-specific. You can search for an exact match of an entire phrase by enclosing the phrase in double quotes. If you do not need an exact match of a phrase, use single terms with OR/AND. Boolean operators are strict. OR is better in most cases. Do not include date constraints in the query: use "start_date" and "end_date" parameters instead. Use Latin script for names. For example, search "Ilya Gusev" instead of "Илья Гусев". Example queries: all:"machine learning" au:"del maestro" au:vaswani AND abs:"attention is all" all:role OR all:playing OR all:"language model" (au:vaswani OR au:"del maestro") ANDNOT ti:attention Args: query: The search query, required. offset: The offset to scroll search results. 10 items will be skipped if offset=10. 0 by default. limit: The maximum number of items to return. limit=5 by default, limit=10 is the maximum. sort_by: 3 options to sort by: relevance, lastUpdatedDate, submittedDate. relevance by default. sort_order: 2 sort orders: ascending, descending. descending by default. include_abstracts: include abstracts in the result or not. False by default. start_date: Start date in %Y-%m-%d format. None by default. end_date: End date in %Y-%m-%d format. None by default.

Downloads a paper from Arxiv and converts it to text. Use mode = "html" by default. Fall back to mode = "pdf" if there are any problems with the HTML version. Args: paper_id: ID of the paper on Arxiv. For instance: 2409.06820v1 include_references: include "references" in the result or not. False by default. mode: Which version of paper to use. Options: ["html", "pdf"]. "html" by default.

Visit a webpage and return the content. Try to use both "tavily" and "basic" providers. They might work differently for the same URL. Args: url: The URL of the webpage to visit. provider: The provider to use. Available providers: "tavily" (default), "exa", or "basic".

Get all papers that cited a given arXiv paper based on Semantic Scholar info. Args: arxiv_id: The ID of a given arXiv paper. offset: The offset to scroll through citations. 10 items will be skipped if offset=10. 0 by default. limit: The maximum number of items to return. limit=50 by default.

Get all papers that were cited by a given arXiv paper (references) based on Semantic Scholar info. Args: arxiv_id: The ID of a given arXiv paper. offset: The offset to scroll through citations. 10 items will be skipped if offset=10. 0 by default. limit: The maximum number of items to return. limit=50 by default.