SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.09665 · ACL · 2022

Automated Crossword Solving

Dan Klein, Kevin Yang, Eric Wallace, Eshaan Pathak, Nicholas Tomlin, Albert Xu, Matthew Ginsberg

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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; 1 failure of our instrument; 3 no recorded run), measured 2026-11-22; 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/2205.09665.json · badge: /api/ran/2205.09665.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
albertkx/berkeley-crossword-solver canonical (classifier label)
classifier basis not recorded on the link
3 ran · 3 no recorded run
FunctionStatusWhere it lives
diff Ran
ran on a synthesized input
albertkx/berkeley-crossword-solver/generate_demo_states.py
code served (permissive licence) · get_code("039616d0b2ba9038")
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 2, 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

# The function diff expects two states, which appear to be 2D grids/lists
# where each cell contains a character. The function compares adjacent states
# to find changed anchor points, treating '.' as a boundary marker and ' ' as empty.

# Creating two simple grid states for testing
# state1: a 3x4 grid at time t-1
state1 = [
    ['a', 'b', 'c', 'd'],
    ['e', 'f', 'g', 'h'],
    ['i', 'j', 'k', 'l']
]

# state2: a 3x4 grid at time t with some differences
state2 = [
    ['a', 'b', 'c', 'd'],
    ['e', 'f', 'x', 'h'],  # 'g' changed to 'x'
    ['i', 'j', 'k', 'l']
]

ARGS = {
    'state1': state1,
    'state2': state2
}
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 2, 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
fill_black_squares Ran
ran on a synthesized input
albertkx/berkeley-crossword-solver/generate_demo_states.py
code served (permissive licence) · get_code("9e0e891f08ffdff8")
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 2, 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.

from copy import deepcopy

# Create sample grids for the function
# solver_grid: a 2D list/array that the function will modify
# ref_grid: a 2D reference grid used to determine which cells should be black ('.')

solver_grid = [
    ['A', 'B', 'C'],
    ['D', '', 'F'],
    ['G', 'H', 'I']
]

ref_grid = [
    ['A', 'B', 'C'],
    ['D', '.', 'F'],
    ['G', 'H', 'I']
]

ARGS = {
    "solver_grid": solver_grid,
    "ref_grid": ref_grid
}
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 list, ndim 2, 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
weighted_avg Ran
ran; our contract draft was wrong, not the code
albertkx/berkeley-crossword-solver/DrQA/drqa/reader/layers.py
code served (permissive licence) · get_code("880a4e01d58b0591")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 torch

torch.manual_seed(0)

# x: batch * len * hdim
batch_size = 2
seq_len = 3
hdim = 4

x = torch.randn(batch_size, seq_len, hdim)

# weights: batch * len, sum(dim=1) = 1
weights = torch.softmax(torch.randn(batch_size, seq_len), dim=1)

ARGS = {
    "x": x,
    "weights": weights,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
setup_t5_reranker Not yet run
unverified (harvested, no recorded run)
albertkx/berkeley-crossword-solver/models.py
code served (permissive licence) · get_code("add8220331a0e044")
How it ran

no invocation recorded.

t5_reranker_score_with_clue Not yet run
unverified (harvested, no recorded run)
albertkx/berkeley-crossword-solver/models.py
code served (permissive licence) · get_code("b24b5669cd475bc4")
How it ran

no invocation recorded.

uniform_weights Not yet run
unverified (harvested, no recorded run)
albertkx/berkeley-crossword-solver/DrQA/drqa/reader/layers.py
code served (permissive licence) · get_code("d5761bf2b769f3cf")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 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.

FunctionAlso present inEarliest-datedOthers named
weighted_avg 3 other papers
3 dated earlier
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1905.05733 (2019-05)
Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering
2021.acl-long.164 (2021 (year only))
BERTAC: Enhancing Transformer-based Language Models with Adversarially Pretrained Convolutional Neural Networks

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

We present the Berkeley Crossword Solver, a state-of-the-art approach for automatically solving crossword puzzles. Our system works by generating answer candidates for each crossword clue using neural question answering models and then combines loopy belief propagation with local search to find full puzzle solutions. Compared to existing approaches, our system improves exact puzzle accuracy from 71% to 82% on crosswords from The New York Times and obtains 99.9% letter accuracy on themeless puzzles. Additionally, in 2021, a hybrid of our system and the existing Dr.Fill system outperformed all human competitors for the first time at the American Crossword Puzzle Tournament. To facilitate research on question answering and crossword solving, we analyze our system's remaining errors and release a dataset of over six million question-answer pairs. Category Clue Answer QA Recall Knowledge (37%) Birds on Minnesota state quarters LOONS ✔ Architect Frank GEHRY ✔ Definition (33%) First in a series PILOT ✔ Tusked savanna dweller WARTHOG ✔ Commonsense (14%) Like games decided by buzzer beaters CLOSE ✔ Opposite of luego AHORA ✔ Wordplay (8%) Frequent book setting SHELF

For agents

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

get_harvested_code_for_paper("2205.09665")
get_code_for_paper("2205.09665")
have("2205.09665")

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

curl https://syntology.ai/api/ran/2205.09665.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.09665.svg)](https://syntology.ai/paper/2205.09665)

Connect an agent — have() is free.