SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.05080 · ICLR · 2025

ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery

Huan Sun, Botao Yu, Yuting Ning, Yu Su, Chen Wei, Shijie Chen, Boshi Wang, Yifei Li, Song Gao, Ziru Chen, Xia Ning, Zeyi Liao, and 8 more

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 0 have no contract verdict; 2 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/2410.05080.json · badge: /api/ran/2410.05080.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
osu-nlp-group/scienceagentbench canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
format_task_dict Ran
ran; our contract draft was wrong, not the code
osu-nlp-group/scienceagentbench/run_infer.py
code served (permissive licence) · get_code("fdaf926f78110e07")
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) and returned dict. 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 torch
import numpy as np

class Args:
    def __init__(self):
        self.datasets_path = "/data/datasets/"
        self.use_knowledge = True

example = {
    'task_inst': 'Classify the iris dataset',
    'dataset_folder_tree': '    iris/\n        data.csv\n        metadata.json',
    'dataset_preview': 'sepal_length,sepal_width,petal_length,petal_width,species\n5.1,3.5,1.4,0.2,setosa',
    'output_fname': 'iris_classification_output.json',
    'domain_knowledge': 'Iris flowers have 4 measurements and 3 species classes'
}

args = Args()

ARGS = {
    'example': example,
    'args': args
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_gold_predictions Ran
ran; our contract draft was wrong, not the code
osu-nlp-group/scienceagentbench/evaluation/harness/run_evaluation.py
code served (permissive licence) · get_code("5d5970aebb56fef0")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 torch
import numpy as np
from unittest.mock import Mock, patch

torch.manual_seed(0)

# Define mock dataset structure
KEY_INSTANCE_ID = 'instance_id'

# Create mock dataset with the required structure
mock_dataset = [
    {KEY_INSTANCE_ID: 'instance_001', 'text': 'sample data 1'},
    {KEY_INSTANCE_ID: 'instance_002', 'text': 'sample data 2'},
]

# Create a mock function to replace load_scienceagentbench_dataset
def mock_load_scienceagentbench_dataset(dataset_name: str, split: str):
    return mock_dataset

# Patch the load_scienceagentbench_dataset function in the module
import sys
current_module = sys.modules[__name__]
current_module.load_scienceagentbench_dataset = mock_load_scienceagentbench_dataset

# Also add it as a builtin for the function to access
import builtins
builtins.load_scienceagentbench_dataset = mock_load_scienceagentbench_dataset
builtins.KEY_INSTANCE_ID = KEY_INSTANCE_ID

ARGS = {
    'dataset_name': 'test_dataset',
    'split': 'train',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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

The advancements of large language models (LLMs) have piqued growing interest in developing LLM-based language agents to automate scientific discovery endto-end, which has sparked both excitement and skepticism about their true capabilities. In this work, we call for rigorous assessment of agents on individual tasks in a scientific workflow before making bold claims on end-to-end automation. To this end, we present ScienceAgentBench, a new benchmark for evaluating language agents for data-driven scientific discovery. To ensure the scientific authenticity and real-world relevance of our benchmark, we extract 102 tasks from 44 peerreviewed publications in four disciplines and engage nine subject matter experts to validate them. We unify the target output for every task to a self-contained Python program file and employ an array of evaluation metrics to examine the generated programs, execution results, and costs. Each task goes through multiple rounds of manual validation by annotators and subject matter experts to ensure its annotation quality and scientific plausibility. We also propose two effective strategies to mitigate data contamination concerns. Using ScienceAgentBench, we evaluate five open-weight and proprietary LLMs, each with three frameworks: direct prompting, OpenHands CodeAct, and self-debug. Given three attempts for each task, the best-performing agent can only solve 32.4% of the tasks independently and 34.3% with expert-provided knowledge. In addition, we evaluate OpenAI o1preview with direct prompting and self-debug, which can boost the performance to 42.2%, demonstrating the effectiveness of increasing inference-time compute but with more than 10 times the cost of other LLMs. Still, our results underscore the limitations of current language agents in generating code for data-driven discovery, let alone end-to-end automation for scientific research.

For agents

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

get_harvested_code_for_paper("2410.05080")
get_code_for_paper("2410.05080")
have("2410.05080")

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

curl https://syntology.ai/api/ran/2410.05080.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.05080.svg)](https://syntology.ai/paper/2410.05080)

Connect an agent — have() is free.