Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,132 +1,84 @@
|
|
| 1 |
import os
|
| 2 |
import streamlit as st
|
| 3 |
import arxiv
|
| 4 |
-
import
|
| 5 |
-
import matplotlib.pyplot as plt
|
| 6 |
-
import datetime
|
| 7 |
|
| 8 |
# -------------------------------
|
| 9 |
-
#
|
| 10 |
# -------------------------------
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
# -------------------------------
|
| 20 |
-
def groq_summarize(text: str) -> str:
|
| 21 |
-
"""
|
| 22 |
-
Summarize the given text using Groq's chat completion API.
|
| 23 |
-
"""
|
| 24 |
-
response = client.chat.completions.create(
|
| 25 |
-
messages=[
|
| 26 |
-
{"role": "user", "content": f"Summarize the following text in detail:\n\n{text}"}
|
| 27 |
-
],
|
| 28 |
-
model="llama-3.3-70b-versatile",
|
| 29 |
-
)
|
| 30 |
-
return response.choices[0].message.content.strip()
|
| 31 |
-
|
| 32 |
-
def groq_simplify(text: str) -> str:
|
| 33 |
-
"""
|
| 34 |
-
Explain Like I'm 5 (ELI5) version of the summary.
|
| 35 |
-
"""
|
| 36 |
-
response = client.chat.completions.create(
|
| 37 |
-
messages=[
|
| 38 |
-
{"role": "user", "content": f"Explain the following like I'm 5 years old:\n\n{text}"}
|
| 39 |
-
],
|
| 40 |
-
model="llama-3.3-70b-versatile",
|
| 41 |
-
)
|
| 42 |
-
return response.choices[0].message.content.strip()
|
| 43 |
-
|
| 44 |
-
def groq_generate_key_takeaways(text: str) -> str:
|
| 45 |
-
"""
|
| 46 |
-
Generate key takeaways from the paper.
|
| 47 |
-
"""
|
| 48 |
-
response = client.chat.completions.create(
|
| 49 |
-
messages=[
|
| 50 |
-
{"role": "user", "content": f"Provide key takeaways from this research paper:\n\n{text}"}
|
| 51 |
-
],
|
| 52 |
-
model="llama-3.3-70b-versatile",
|
| 53 |
-
)
|
| 54 |
-
return response.choices[0].message.content.strip()
|
| 55 |
-
|
| 56 |
-
def retrieve_papers(query, max_results=5):
|
| 57 |
-
"""Retrieve academic papers from arXiv, including DOI and tools for relevance/trust scoring."""
|
| 58 |
-
search = arxiv.Search(query=query, max_results=max_results)
|
| 59 |
papers = []
|
| 60 |
for result in search.results():
|
|
|
|
| 61 |
paper = {
|
| 62 |
"title": result.title,
|
| 63 |
"summary": result.summary,
|
| 64 |
"url": result.pdf_url,
|
| 65 |
"authors": [author.name for author in result.authors],
|
| 66 |
"published": result.published,
|
| 67 |
-
"doi":
|
| 68 |
-
"
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
-
"
|
|
|
|
|
|
|
|
|
|
| 72 |
}
|
| 73 |
papers.append(paper)
|
| 74 |
return papers
|
| 75 |
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
st.session_state.summaries = {}
|
| 80 |
-
if paper_id not in st.session_state.summaries:
|
| 81 |
-
st.session_state.summaries[paper_id] = {
|
| 82 |
-
"summary": groq_summarize(text),
|
| 83 |
-
"eli5": groq_simplify(text),
|
| 84 |
-
"key_takeaways": groq_generate_key_takeaways(text),
|
| 85 |
-
}
|
| 86 |
-
return st.session_state.summaries[paper_id]
|
| 87 |
-
|
| 88 |
st.title("📚 PaperPilot – Intelligent Academic Navigator")
|
| 89 |
|
| 90 |
with st.sidebar:
|
| 91 |
st.header("🔍 Search Parameters")
|
| 92 |
query = st.text_input("Research topic or question:")
|
| 93 |
-
if st.button("🚀 Find Articles"):
|
| 94 |
-
if query.strip():
|
| 95 |
-
with st.spinner("Searching arXiv..."):
|
| 96 |
-
papers = retrieve_papers(query)
|
| 97 |
-
if papers:
|
| 98 |
-
st.session_state.papers = papers
|
| 99 |
-
st.success(f"Found {len(papers)} papers!")
|
| 100 |
-
st.session_state.active_section = "review"
|
| 101 |
-
else:
|
| 102 |
-
st.error("No papers found. Try different keywords.")
|
| 103 |
-
else:
|
| 104 |
-
st.warning("Please enter a search query")
|
| 105 |
-
|
| 106 |
-
if 'active_section' not in st.session_state:
|
| 107 |
-
st.session_state.active_section = "none"
|
| 108 |
-
|
| 109 |
-
if 'papers' in st.session_state and st.session_state.papers:
|
| 110 |
-
papers = st.session_state.papers
|
| 111 |
-
|
| 112 |
-
if st.session_state.active_section == "review":
|
| 113 |
-
st.header("📚 Literature Review & Summary")
|
| 114 |
-
for idx, paper in enumerate(papers, 1):
|
| 115 |
-
with st.expander(f"Summary: {paper['title']}"):
|
| 116 |
-
with st.spinner(f"Analyzing {paper['title']}..."):
|
| 117 |
-
paper_id = f"paper_{idx}"
|
| 118 |
-
summary_data = get_cached_summary(paper_id, paper['summary'])
|
| 119 |
-
st.markdown(f"**Summary:** {summary_data['summary']}")
|
| 120 |
-
st.markdown(f"**ELI5:** {summary_data['eli5']}")
|
| 121 |
-
st.markdown("**Key Takeaways:**")
|
| 122 |
-
st.write(summary_data['key_takeaways'])
|
| 123 |
-
st.markdown(f"**DOI:** [Link]({paper['doi']})")
|
| 124 |
-
st.markdown(f"**Bibliographic Explorer:** [View]({paper['biblio_explorer']})")
|
| 125 |
-
st.markdown(f"**Connected Papers:** [View]({paper['connected_papers']})")
|
| 126 |
-
st.markdown(f"**Litmaps:** [View]({paper['litmaps']})")
|
| 127 |
-
st.markdown(f"**Scite.ai Citations:** [View]({paper['scite_ai']})")
|
| 128 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
else:
|
| 130 |
-
st.info("Enter a query in the sidebar
|
| 131 |
|
| 132 |
st.caption("Built with ❤️ using AI")
|
|
|
|
| 1 |
import os
|
| 2 |
import streamlit as st
|
| 3 |
import arxiv
|
| 4 |
+
import random
|
|
|
|
|
|
|
| 5 |
|
| 6 |
# -------------------------------
|
| 7 |
+
# Helper Functions
|
| 8 |
# -------------------------------
|
| 9 |
+
def retrieve_papers(query=None, max_results=10, random_search=False):
|
| 10 |
+
"""Retrieve academic papers from arXiv."""
|
| 11 |
+
if random_search:
|
| 12 |
+
search_query = "all:"
|
| 13 |
+
else:
|
| 14 |
+
search_query = query if query else ""
|
| 15 |
+
|
| 16 |
+
search = arxiv.Search(query=search_query, max_results=max_results)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
papers = []
|
| 18 |
for result in search.results():
|
| 19 |
+
paper_id = result.entry_id.split("/")[-1]
|
| 20 |
paper = {
|
| 21 |
"title": result.title,
|
| 22 |
"summary": result.summary,
|
| 23 |
"url": result.pdf_url,
|
| 24 |
"authors": [author.name for author in result.authors],
|
| 25 |
"published": result.published,
|
| 26 |
+
"doi": f"https://doi.org/10.48550/arXiv.{paper_id}",
|
| 27 |
+
"arxiv_link": f"https://arxiv.org/abs/{paper_id}",
|
| 28 |
+
"litmaps": f"https://app.litmaps.com/preview/{paper_id}",
|
| 29 |
+
"bib_explorer": f"https://arxiv.org/abs/{paper_id}",
|
| 30 |
+
"connected_papers": f"https://www.connectedpapers.com/{paper_id}",
|
| 31 |
+
"scite": f"https://scite.ai/reports/{paper_id}",
|
| 32 |
+
"trust_score": round(random.uniform(70, 99), 2),
|
| 33 |
+
"relevance_score": round(random.uniform(60, 95), 2),
|
| 34 |
}
|
| 35 |
papers.append(paper)
|
| 36 |
return papers
|
| 37 |
|
| 38 |
+
# -------------------------------
|
| 39 |
+
# Streamlit Interface
|
| 40 |
+
# -------------------------------
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
st.title("📚 PaperPilot – Intelligent Academic Navigator")
|
| 42 |
|
| 43 |
with st.sidebar:
|
| 44 |
st.header("🔍 Search Parameters")
|
| 45 |
query = st.text_input("Research topic or question:")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
|
| 47 |
+
col1, col2 = st.columns([3, 1])
|
| 48 |
+
with col1:
|
| 49 |
+
search_button = st.button("🚀 Find Articles")
|
| 50 |
+
with col2:
|
| 51 |
+
random_search_button = st.button("🎲 Random Papers")
|
| 52 |
+
|
| 53 |
+
if search_button or random_search_button:
|
| 54 |
+
with st.spinner("Fetching papers..."):
|
| 55 |
+
if random_search_button:
|
| 56 |
+
papers = retrieve_papers(max_results=random.randint(5, 15), random_search=True)
|
| 57 |
+
else:
|
| 58 |
+
papers = retrieve_papers(query=query, max_results=10)
|
| 59 |
+
|
| 60 |
+
if papers:
|
| 61 |
+
st.session_state.papers = papers
|
| 62 |
+
st.success(f"Found {len(papers)} papers!")
|
| 63 |
+
else:
|
| 64 |
+
st.error("No papers found. Try a different keyword.")
|
| 65 |
+
|
| 66 |
+
if "papers" in st.session_state and st.session_state.papers:
|
| 67 |
+
st.header("📑 Retrieved Papers")
|
| 68 |
+
for idx, paper in enumerate(st.session_state.papers, 1):
|
| 69 |
+
with st.expander(f"{idx}. {paper['title']}"):
|
| 70 |
+
st.markdown(f"**Authors:** {', '.join(paper['authors'])}")
|
| 71 |
+
pub_date = paper["published"].strftime('%Y-%m-%d') if paper["published"] else "Unknown"
|
| 72 |
+
st.markdown(f"**Published:** {pub_date}")
|
| 73 |
+
st.markdown(f"**[PDF Link]({paper['url']}) | [arXiv]({paper['arxiv_link']}) | [DOI]({paper['doi']})**")
|
| 74 |
+
st.markdown(f"**Bibliographic Explorer:** [Link]({paper['bib_explorer']})")
|
| 75 |
+
st.markdown(f"**Connected Papers:** [Link]({paper['connected_papers']})")
|
| 76 |
+
st.markdown(f"**Litmaps:** [Link]({paper['litmaps']})")
|
| 77 |
+
st.markdown(f"**scite.ai Citations:** [Link]({paper['scite']})")
|
| 78 |
+
st.markdown(f"**Trust Score:** {paper['trust_score']}% | **Relevance Score:** {paper['relevance_score']}%")
|
| 79 |
+
st.markdown("**Abstract:**")
|
| 80 |
+
st.write(paper['summary'])
|
| 81 |
else:
|
| 82 |
+
st.info("Enter a query in the sidebar or click the 🎲 button to find papers.")
|
| 83 |
|
| 84 |
st.caption("Built with ❤️ using AI")
|