neubig

import requests
import sys
import time

sleep_time = 20
def query_api(url, session):
  global sleep_time
  time.sleep(sleep_time / 1000.0)
  r = session.get(url)
  while r.status_code == 429:

neubig

# NOTE:
# You can find an updated, more robust and feature-rich implementation
# in Zeno Build
# - Zeno Build: https://github.com/zeno-ml/zeno-build/
# - Implementation: https://github.com/zeno-ml/zeno-build/blob/main/zeno_build/models/providers/openai_utils.py

import openai
import asyncio
from typing import Any

neubig

OpenScholar is a retrieval-augmented language model that assists researchers in synthesizing scientific literature. The system uses a database of 45 million open-access papers to provide citation-backed responses to queries, accurately identifying relevant passages and generating reliable answers across multiple scientific domains. This approach addresses the growing challenge of keeping up with rapidly expanding scientific literature.

The researchers developed ScholarQABench, a multi-domain benchmark for evaluating literature search capabilities, with 2,967 expert-written queries and 208 detailed answers across computer science, physics, neuroscience, and biomedicine. In testing, OpenScholar-8B outperformed GPT-4o by 5% and PaperQA2 by 7% in correctness metrics, despite being a smaller, open model.

Citation accuracy stands as a key strength of OpenScholar. While GPT-4o shows concerning citation hallucination rates of 78-90%, OpenScholar matches human expert-level accuracy in citation verification. The syst

neubig

# Export your NAACL 2025 events into ics format
# 1. Go to the NAACL schedule and download it as a csv: https://docs.google.com/spreadsheets/d/1SXIF0ovLudQ4UvR0nTyagDcgnn9zdulhUY578mvQpRk/edit?gid=1679189789#gid=1679189789
# 2. Change MY_NAME below and run the program
# 3. Go to "Settings" in Google Calendar, click Import/Export, and import the file to your calendar

MY_NAME = "Neubig"

import csv
import os
from datetime import datetime

neubig

Running OpenHands GUI on Windows

This guide provides step-by-step instructions for running OpenHands on a Windows machine without using WSL or Docker.

Prerequisites

1. Windows 10/11 - A modern Windows operating system
2. PowerShell 5.1 or PowerShell 7+ - Windows PowerShell comes pre-installed on Windows 10/11, but PowerShell 7+ is recommended for better compatibility
3. Python 3.12 - Python 3.12 is required (Python 3.14 is not supported due to pythonnet compatibility)
4. Git - For cloning the repository and version control

neubig

import os, csv, requests, datetime as dt, calendar, time, random
from collections import OrderedDict
import argparse
import matplotlib.pyplot as plt

BASE_SEARCH_URL = 'https://api.github.com/search/issues'


def parse_link_header(value: str):
    links = {}

neubig

# Get the data from here: https://github.com/martenlienen/icml-neurips-iclr-dataset

import pandas as pd
import re
from collections import Counter
import matplotlib.pyplot as plt

def classify_affiliation(affiliation):
    """Classify an affiliation into university, industry, or other."""
    if pd.isna(affiliation) or affiliation == 'None' or affiliation.strip() == '':

neubig

"""Fusion-style delegation harness built with the OpenHands SDK.

Install:
    uv pip install openhands-sdk openhands-tools

Run:
    export LLM_API_KEY="..."  # or export OPENHANDS_API_KEY="..."
    export MAIN_MODEL="openhands/gpt-5.5"
    export SIDEKICK_MODEL="openhands/minimax-m2.7"
    uv run python fusion_harness_example.py "Find and fix the failing tests in this repo."