Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| import pandas as pd | |
| # Title and description | |
| st.set_page_config(page_title="SLR-BENCH Leaderboard", layout="wide") | |
| st.title("SLR-BENCH Leaderboard") | |
| st.markdown(""" | |
| This is a leaderboard displaying model performance metrics for the SLR-BENCH benchmark. | |
| We report the models’ Logical Reasoning Level (LRL), syntax score, | |
| stage-specific logical reasoning accuracy (basic, easy, medium, hard), total completion tokens, and inference cost. | |
| Higher LRL and accuracy indicate superior logical reasoning; lower compute, greater efficiency. Performance drops | |
| as complexity increases, while Reasoning LLMs consistently outperform conventional LLMs. | |
| """) | |
| # Load the CSV file | |
| csv_file_path = "slr_bench_results.csv" # Update this path to your CSV file | |
| try: | |
| data = pd.read_csv(csv_file_path) | |
| # Display the table | |
| st.dataframe(data, use_container_width=True) | |
| except FileNotFoundError: | |
| st.error(f"CSV file not found at '{csv_file_path}'. Please ensure the file exists.") | |