SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.01717 · EMNLP · 2020

STORIUM: A Dataset and Evaluation Platform for Machine-in-the-Loop Story Generation

Shufan Wang, Nanyun Peng, Mohit Iyyer, Nader Akoury, Josh Whiting, Stephen Hood

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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 4 have no contract verdict; 0 failures of our instrument; 1 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/2010.01717.json · badge: /api/ran/2010.01717.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
dojoteef/storium-frontend canonical (classifier label)
classifier basis: census
4 ran · 1 no recorded run
FunctionStatusWhere it lives
from_db_type Ran
ran on a synthesized input
dojoteef/storium-frontend/src/woolgatherer/db/types.py
code served (permissive licence) · get_code("8e7aff1d4253f338")
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 SampleModel. 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 json
from typing import Any, Dict, Mapping, Union
from pydantic import BaseModel, Json

class SampleModel(BaseModel):
    name: str
    value: int
    description: str = "default"

class AnotherModel(BaseModel):
    count: int
    active: bool = True

# Test case 1: cls is a Pydantic BaseModel, obj is a JSON string
cls_arg = SampleModel
obj_arg = json.dumps({"name": "test", "value": 42, "description": "test description"})

ARGS = {
    "cls": cls_arg,
    "obj": obj_arg
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SampleModel — a SampleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SampleModel 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
json_hash Ran
ran on a synthesized input
dojoteef/storium-frontend/src/woolgatherer/db/utils.py
code served (permissive licence) · get_code("186810128079271e")
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 tuple, 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 json
import hashlib
from functools import partial
from datetime import datetime
from typing import Any, Dict, Tuple

class JSONEncoder(json.JSONEncoder):
    """ A custom JSON encoder which handles datetime objects """

    def default(self, o):
        """ Overloaded to support datetime objects """
        if isinstance(o, datetime):
            return o.isoformat(' ')
        return super().default(o)

json_dumps = partial(json.dumps, cls=JSONEncoder, separators=(',', ':'))

def normalized_json_str(json_obj: Dict[str, Any]) -> str:
    """
    Normalize a JSON object by sorting it and removing any extraneous space.
    """
    return json_dumps(json_obj, sort_keys=True, ensure_ascii=False)

def json_hash(json_obj: Dict[str, Any]) -> Tuple[str, str]:
    """
    Generate a consistent hash for a given JSON object. We need this in order
    to have a consistent hash that we can compute for JSON in the database when
    checking for equality. For example, while PostgreSQL has support for unique
    constraints on JSONB columns, it has a size limit which we will definitely
    go over!
    """
    hasher = hashlib.md5()
    json_str = normalized_json_str(json_obj)
    hasher.update(json_str.encode('utf-8'))
    return (json_str, hasher.hexdigest())


# Construct valid arguments for json_hash function
ARGS = {
    "json_obj": {
        "name": "test_config",
        "version": 1,
        "timestamp": datetime(2024, 1, 15, 10, 30, 45),
        "nested": {
            "key1": "value1",
            "key2": [1, 2, 3],
            "key3": {"deep": "value"}
        },
        "parameters": {
            "learning_rate": 0.001,
            "batch_size": 32,
            "epochs": 100
        }
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple 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
normalized_json_str Ran
ran on a synthesized input
dojoteef/storium-frontend/src/woolgatherer/db/utils.py
code served (permissive licence) · get_code("c5eecd894b0527d1")
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 str. 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 json
from functools import partial
from datetime import datetime
from typing import Any, Dict, Tuple

class JSONEncoder(json.JSONEncoder):
    """ A custom JSON encoder which handles datetime objects """

    def default(self, o):
        """ Overloaded to support datetime objects """
        if isinstance(o, datetime):
            return o.isoformat(' ')
        return super().default(o)

json_dumps = partial(json.dumps, cls=JSONEncoder, separators=(',', ':'))

def normalized_json_str(json_obj: Dict[str, Any]) -> str:
    """
    Normalize a JSON object by sorting it and removing any extraneous space.
    """
    return json_dumps(json_obj, sort_keys=True, ensure_ascii=False)


# Setup: Create valid arguments for normalized_json_str
ARGS = {
    "json_obj": {
        "name": "example",
        "value": 42,
        "timestamp": datetime(2023, 11, 15, 10, 30, 45),
        "nested": {
            "key1": "value1",
            "key2": [1, 2, 3]
        },
        "boolean": True,
        "null_value": None
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
to_db_type Ran
ran on a synthesized input
dojoteef/storium-frontend/src/woolgatherer/db/types.py
code served (permissive licence) · get_code("9840cb33e4b5579d")
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 str. 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 json
import datetime
from typing import Any
from pydantic import BaseModel

class SampleModel(BaseModel):
    name: str
    value: int

# Create test objects for each parameter type
datetime_obj = datetime.datetime(2023, 10, 15, 14, 30, 45)
date_obj = datetime.date(2023, 10, 15)
time_obj = datetime.time(14, 30, 45)
model_obj = SampleModel(name="test", value=42)
regular_obj = "regular string"

# Test with a datetime.datetime object
ARGS = {
    "obj": datetime_obj
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pydantic — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_sa_type Not yet run
unverified (harvested, no recorded run)
dojoteef/storium-frontend/src/woolgatherer/db/types.py
code served (permissive licence) · get_code("3673cc5a808e8c03")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

Systems for story generation are asked to produce plausible and enjoyable stories given an input context. This task is underspecified, as a vast number of diverse stories can originate from a single input. The large output space makes it difficult to build and evaluate story generation models, as (1) existing datasets lack rich enough contexts to meaningfully guide models, and (2) existing evaluations (both crowdsourced and automatic) are unreliable for assessing long-form creative text. To address these issues, we introduce a dataset and evaluation platform built from STORIUM, an online collaborative storytelling community. Our author-generated dataset contains 6K lengthy stories (125M tokens) with fine-grained natural language annotations (e.g., character goals and attributes) interspersed throughout each narrative, forming a robust source for guiding models. We evaluate language models fine-tuned on our dataset by integrating them onto STORIUM, where real authors can query a model for suggested story continuations and then edit them. Automatic metrics computed over these edits correlate well with both user ratings of generated stories and qualitative feedback from semi-structured user interviews. We release both the STORIUM dataset and evaluation platform to spur more principled research into story generation.

For agents

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

get_harvested_code_for_paper("2010.01717")
get_code_for_paper("2010.01717")
have("2010.01717")

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

curl https://syntology.ai/api/ran/2010.01717.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.01717.svg)](https://syntology.ai/paper/2010.01717)

Connect an agent — have() is free.