SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.16702 · 2024

Accurate and Nuanced Open-QA Evaluation Through Textual Entailment

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2405.16702.json · badge: /api/ran/2405.16702.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
U-Alberta/QA-partial-marks canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
make_openai_request Ran
ran on a synthesized input
U-Alberta/QA-partial-marks/llm_tools.py
code served (permissive licence) · get_code("ee7d2f1bfe7b5774")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned coroutine. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import asyncio
import aiohttp
from unittest.mock import AsyncMock, MagicMock, patch
import os

# Set up a minimal environment
os.environ['OPENAI_API_KEY'] = 'sk-test-key-12345'

# Create a mock session that behaves like aiohttp.ClientSession
mock_session = AsyncMock()

# Create a mock response
mock_response = AsyncMock()
mock_response.json = AsyncMock(return_value={
    'choices': [{'message': {'content': 'This is a test response from the assistant.'}}]
})
mock_response.raise_for_status = MagicMock()

# Set up the session.post to return the mock response as an async context manager
mock_session.post = MagicMock()
mock_session.post.return_value.__aenter__ = AsyncMock(return_value=mock_response)
mock_session.post.return_value.__aexit__ = AsyncMock(return_value=None)

# Define arguments for the function
ARGS = {
    'session': mock_session,
    'prompt': 'What is the capital of France?',
    'system': 'You are a helpful geography assistant.',
    'model': 'gpt-3.5-turbo',
    'temperature': 0.7,
    'max_tokens': 150,
    'seed': 42
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
aiohttp — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type coroutine — a coroutine output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a coroutine output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
make_openai_request_batch Ran
ran on a synthesized input
U-Alberta/QA-partial-marks/llm_tools.py
code served (permissive licence) · get_code("8b398f59888ac6d8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned coroutine. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import asyncio
import aiohttp
from unittest.mock import Mock, AsyncMock, patch
import os

# Set up environment variable
os.environ['OPENAI_API_KEY'] = 'test-api-key-12345'

# Create a mock session that behaves like aiohttp.ClientSession
mock_session = AsyncMock(spec=aiohttp.ClientSession)

# Create mock response context manager
mock_response = AsyncMock()
mock_response.json = AsyncMock(return_value={
    'choices': [{'message': {'content': 'Test response from API'}}]
})
mock_response.raise_for_status = Mock()

# Mock the post method to return an async context manager
mock_session.post = Mock(return_value=AsyncMock(
    __aenter__=AsyncMock(return_value=mock_response),
    __aexit__=AsyncMock(return_value=None)
))

# Define arguments for make_openai_request_batch
ARGS = {
    'session': mock_session,
    'prompts': [
        'What is machine learning?',
        'Explain deep learning.',
        'What is a neural network?'
    ],
    'system': 'You are an expert in artificial intelligence.',
    'model': 'gpt-3.5-turbo',
    'temperature': 0.5,
    'max_tokens': 150
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
aiohttp — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type coroutine — a coroutine output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a coroutine output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Open-domain question answering (Open-QA) is a common task for evaluating large language models (LLMs). However, current Open-QA evaluations are criticized for the ambiguity in questions and the lack of semantic understanding in evaluators. Complex evaluators, powered by foundation models or LLMs and pertaining to semantic equivalence, still deviate from human judgments by a large margin. We propose to study the entailment relations of answers to identify more informative and more general system answers, offering a much closer evaluation to human judgment on both NaturalQuestions and TriviaQA while being learning-free. The entailment-based evaluation we propose allows the assignment of bonus or partial marks by quantifying the inference gap between answers, enabling a nuanced ranking of answer correctness that has higher AUC than current methods.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2405.16702")
get_code_for_paper("2405.16702")
have("2405.16702")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2405.16702.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2405.16702.svg)](https://syntology.ai/paper/2405.16702)

Connect an agent — have() is free.