Implementation:ThreeSR Awesome Inference Time Scaling Get Paper Info Function

Page Metadata
Knowledge Sources	Awesome-Inference-Time-Scaling
Domains	Metadata Enrichment, API Integration, Scholarly Literature Discovery
Last Updated	2026-02-14 00:00 GMT

Overview

Concrete tool for fetching extended paper metadata from the Semantic Scholar API by paper ID, provided by the get_paper_info() function in fetch_semantic_info.py.

Description

The get_paper_info() function takes a Semantic Scholar paper ID (obtained from search results) and queries the Semantic Scholar v1 paper detail endpoint to retrieve extended metadata. This enriches the basic search result with additional fields -- most importantly the arxivId (used to construct arXiv links) and the abstract (displayed in the curated paper list).

Important note: This function uses the older v1 endpoint (https://api.semanticscholar.org/v1/paper/{paper_id}), not the graph/v1 endpoint used by search_papers(). The v1 endpoint returns a different response schema with fields like arxivId at the top level.

Usage

Import or call this function when:

• You have a paper ID from a search result and need the arXiv identifier to build PDF links
• You need the full abstract of a paper for display purposes
• You are enriching search results before formatting them into the README

Code Reference

Source Location: fetch_semantic_info.py, lines 43-46

Function Signature:

def get_paper_info(paper_id: str) -> dict

Import Statement:

from fetch_semantic_info import get_paper_info

Source Code:

def get_paper_info(paper_id):
    url = f'https://api.semanticscholar.org/v1/paper/{paper_id}'
    response = requests.get(url)
    return response.json()

Dependencies:

• requests (third-party HTTP library)

API Endpoint:

GET https://api.semanticscholar.org/v1/paper/{paper_id}

Note: This is the v1 endpoint, distinct from the graph/v1 endpoint used by search_papers().

I/O Contract

Inputs:

Parameter	Type	Default	Description
paper_id	str	(required)	The Semantic Scholar unique paper identifier, typically obtained from the paperId field in search results

Outputs:

Field	Type	Description
arxivId	str or None	The arXiv identifier (e.g., "2501.12345"); used to construct arXiv abs and PDF URLs
abstract	str or None	The full text of the paper's abstract
title	str	The paper title
authors	list[dict]	List of author objects with name and authorId fields
venue	str	The publication venue
year	int	The publication year
citationCount	int	Number of citations (available in v1 response)
references	list[dict]	List of referenced papers (available in v1 response)

The function returns a dict containing the full v1 API response. The fields most commonly used by this repository are arxivId and abstract.

Usage Examples

Example 1: Basic detail retrieval

from fetch_semantic_info import get_paper_info

# Using a Semantic Scholar paper ID from search results
paper_id = "649def34f8be52c8b66281af98ae884c09aef38b"
info = get_paper_info(paper_id)

print(f"Title: {info.get('title')}")
print(f"arXiv ID: {info.get('arxivId')}")
print(f"Abstract: {info.get('abstract', 'No abstract available.')[:200]}...")

Example 2: Constructing arXiv links from the result

from fetch_semantic_info import get_paper_info

paper_id = "649def34f8be52c8b66281af98ae884c09aef38b"
info = get_paper_info(paper_id)

arxiv_id = info.get("arxivId", "N/A")
if arxiv_id and arxiv_id != "N/A":
    abs_url = f"https://arxiv.org/abs/{arxiv_id}"
    pdf_url = f"https://arxiv.org/pdf/{arxiv_id}"
    print(f"Abstract page: {abs_url}")
    print(f"PDF download:  {pdf_url}")
else:
    print("No arXiv ID available for this paper.")

Example 3: Full pipeline -- search then enrich

from fetch_semantic_info import search_papers, get_paper_info

# Step 1: Search for papers
papers = search_papers("Inference-Time Scaling", limit=3)

# Step 2: Enrich each result with detail info
for paper in papers:
    paper_id = paper["paperId"]
    detail = get_paper_info(paper_id)

    title = paper["title"]
    arxiv_id = detail.get("arxivId", "N/A")
    abstract = detail.get("abstract", "No abstract available.")

    print(f"Title: {title}")
    print(f"arXiv: {arxiv_id}")
    print(f"Abstract: {abstract[:150]}...")
    print()

Example 4: Handling missing fields

from fetch_semantic_info import get_paper_info

info = get_paper_info("some-paper-id")

# Safely extract with defaults (as done in format_paper_info)
arxiv_id = info.get("arxivId", "N/A")
abstract = info.get("abstract", "No abstract available.")

print(f"arXiv ID: {arxiv_id}")
print(f"Abstract: {abstract}")

Related Pages

• Principle:ThreeSR_Awesome_Inference_Time_Scaling_Paper_Detail_Retrieval -- The principle of Paper Detail Retrieval that this function implements
• Environment:ThreeSR_Awesome_Inference_Time_Scaling_Python_Runtime_Environment
• Environment:ThreeSR_Awesome_Inference_Time_Scaling_Semantic_Scholar_API_Environment
• Heuristic:ThreeSR_Awesome_Inference_Time_Scaling_API_Rate_Limiting_Tip