SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1711.00482 · 2017

Learning with Latent Language

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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; 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/1711.00482.json · badge: /api/ran/1711.00482.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
jacobandreas/l3 canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
flatten Ran
ran on a synthesized input
jacobandreas/l3/misc/util.py
code served (permissive licence) · get_code("56b9de9f5cc2133a")
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 int; 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

torch.manual_seed(0)

# The flatten function recursively flattens nested lists/tuples into a single list
# It takes one parameter: lol (list of lists or similar nested structure)

# Test with a nested list/tuple structure
nested_structure = [
    [1, 2, 3],
    (4, 5),
    [
        [6, 7],
        8
    ],
    9
]

ARGS = {
    "lol": nested_structure
}
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 int — 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
pad_slice Ran
ran on a synthesized input
jacobandreas/l3/misc/array.py
code served (permissive licence) · get_code("04d4777b58b16f4b")
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 ndarray, ndim 3, dtype family float, shape (4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np

# Create a small 3D numpy array with distinct dimensions
# Shape: (height=5, width=6, channels=3)
array = np.random.rand(5, 6, 3).astype(np.float32)

# Define slice ranges for rows and columns
# slice_r is (r1, r2) where r2 > r1
# slice_c is (c1, c2) where c2 > c1
# Using negative r1 to test padding behavior
slice_r = (-1, 3)  # Will require padding on top
slice_c = (2, 5)   # Normal slice within bounds

ARGS = {
    "array": array,
    "slice_r": slice_r,
    "slice_c": slice_c,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 3, 3) — shape (4, 3, 3) as a later run recorded it
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
parse_fexp Ran
ran on a synthesized input
jacobandreas/l3/misc/util.py
code served (permissive licence) · get_code("bd4ddcaa3bde00a9")
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 re

FEXP_RE = re.compile('(.*)\\[(.*)\\]')

def parse_fexp(fexp):
    m = FEXP_RE.match(fexp)
    return (m.group(1), m.group(2))

# The function takes a single parameter 'fexp' which is a string
# It matches against a regex pattern of the form "something[something_else]"
# and returns a tuple of the two captured groups

ARGS = {
    'fexp': 'layer_name[index_value]'
}
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
tree_map Ran
ran on a synthesized input
jacobandreas/l3/misc/util.py
code served (permissive licence) · get_code("bef84cd330780883")
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. 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

torch.manual_seed(0)

def simple_function(x):
    """A simple function that processes tree nodes"""
    if isinstance(x, (int, float)):
        return x * 2
    elif isinstance(x, torch.Tensor):
        return x + 1
    elif isinstance(x, str):
        return x.upper()
    else:
        return x

# Create a tree structure with mixed types
# A tuple tree where the first element is the "head" and rest are "subtrees"
tree_structure = (
    42,  # head (will be processed)
    (10, 5, 3),  # subtree 1
    (torch.tensor([1.0, 2.0]), torch.tensor([3.0, 4.0])),  # subtree 2
    ("hello", "world"),  # subtree 3
)

ARGS = {
    "function": simple_function,
    "tree": tree_structure,
}
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 tuple — a tuple output has no array shape of its own; none is 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 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 0 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.

FunctionAlso present inEarliest-datedOthers named
flatten 2 other papers
1 dated earlier, 1 later
1601.01705 (2016-01)
Learning to Compose Neural Networks for Question Answering
2506.07255 (2025-06)
Subgoal-Guided Policy Heuristic Search with Learned Subgoals
tree_map 2 other papers
1 dated earlier, 1 later
1601.01705 (2016-01)
Learning to Compose Neural Networks for Question Answering
2506.07255 (2025-06)
Subgoal-Guided Policy Heuristic Search with Learned Subgoals
pad_slice 1 other paper
1 later
2506.07255 (2025-06)
Subgoal-Guided Policy Heuristic Search with Learned Subgoals
parse_fexp 1 other paper
1 later
2506.07255 (2025-06)
Subgoal-Guided Policy Heuristic Search with Learned Subgoals

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 named concepts and compositional operators present in natural language provide a rich source of information about the kinds of abstractions humans use to navigate the world. Can this linguistic background knowledge improve the generality and efficiency of learned classifiers and control policies? This paper aims to show that using the space of natural language strings as a parameter space is an effective way to capture natural task structure. In a pretraining phase, we learn a language interpretation model that transforms inputs (e.g. images) into outputs (e.g. labels) given natural language descriptions. To learn a new concept (e.g. a classifier), we search directly in the space of descriptions to minimize the interpreter's loss on training examples. Crucially, our models do not require language data to learn these concepts: language is used only in pretraining to impose structure on subsequent learning. Results on image classification, text editing, and reinforcement learning show that, in all settings, models with a linguistic parameterization outperform those without.

For agents

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

get_harvested_code_for_paper("1711.00482")
get_code_for_paper("1711.00482")
have("1711.00482")

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

curl https://syntology.ai/api/ran/1711.00482.json

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

[![Syntology run record](https://syntology.ai/api/ran/1711.00482.svg)](https://syntology.ai/paper/1711.00482)

Connect an agent — have() is free.