SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.14073 · 2024

Improving Language Understanding from Screenshots

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/2402.14073.json · badge: /api/ran/2402.14073.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
princeton-nlp/ptp canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
find_module Ran
ran; our contract draft was wrong, not the code
princeton-nlp/ptp/modeling/modeling_ptp.py
code served (permissive licence) · get_code("afbc39b92210ad29")
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 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 torch
from torch import nn

torch.manual_seed(0)

# Create a nested module structure to test find_module
class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Linear(3, 2)

class NestedModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.simple = SimpleModule()
        self.conv = nn.Conv1d(4, 3, 1)

class RootModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.nested = NestedModule()
        self.linear = nn.Linear(4, 2)

# Instantiate the root module
root_module = RootModule()

# Test case: access a deeply nested module using dot notation
# This will traverse: root_module -> nested -> simple -> layer1
key = "nested.simple.layer1"

ARGS = {
    "root_module": root_module,
    "key": key
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — 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
text_forward_flash_attn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
princeton-nlp/ptp/modeling/modeling_ptp.py
code served (permissive licence) · get_code("cc8a9e90ec696f4f")
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 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 torch.nn as nn

torch.manual_seed(0)

# Build a minimal self-attention module that satisfies the function's requirements
class MinimalAttentionModule(nn.Module):
    def __init__(self, hidden_dim=4, n_heads=2, key_value_proj_dim=2, dropout=0.1):
        super().__init__()
        self.hidden_dim = hidden_dim
        self.n_heads = n_heads
        self.key_value_proj_dim = key_value_proj_dim
        self.dropout = dropout
        
        # Linear projections for query, key, value
        self.query = nn.Linear(hidden_dim, n_heads * key_value_proj_dim)
        self.key = nn.Linear(hidden_dim, n_heads * key_value_proj_dim)
        self.value = nn.Linear(hidden_dim, n_heads * key_value_proj_dim)
        
        # Output projection
        self.output = nn.Linear(n_heads * key_value_proj_dim, hidden_dim)
        
        self.training = False

# Create the module instance
module = MinimalAttentionModule(hidden_dim=4, n_heads=2, key_value_proj_dim=2, dropout=0.1)

# Create input tensors
batch_size = 2
seq_length = 3
hidden_dim = 4
n_heads = 2
key_value_proj_dim = 2

hidden_states = torch.randn(batch_size, seq_length, hidden_dim)
mask = None
key_value_states = None
position_bias = None
past_key_value = None
layer_head_mask = None
query_length = None
use_cache = False
output_attentions = False

ARGS = {
    'self': module,
    'hidden_states': hidden_states,
    'mask': mask,
    'key_value_states': key_value_states,
    'position_bias': position_bias,
    'past_key_value': past_key_value,
    'layer_head_mask': layer_head_mask,
    'query_length': query_length,
    'use_cache': use_cache,
    'output_attentions': output_attentions,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
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 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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 1 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
find_module 8 other papers
3 dated earlier, 2 the same month, 3 later
2305.17333 (2023-05)
Fine-Tuning Language Models with Just Forward Passes
2310.09639 (2023-10)
DPZero: Private Fine-Tuning of Language Models without Backpropagation
2312.15184 (2023-12)
ZO-AdaMU Optimizer: Adapting Perturbation by the Momentum and Uncertainty in Zeroth-order Optimization
and 5 more not named here

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

An emerging family of language models (LMs), capable of processing both text and images within a single visual view, has the promise to unlock complex tasks such as chart understanding and UI navigation. We refer to these models as screenshot language models. Despite their appeal, existing screenshot LMs substantially lag behind text-only models on language understanding tasks. To close this gap, we adopt a simplified setting where the model inputs are plain-text-rendered screenshots, and we focus on improving the text ability of screenshot LMs. We propose a novel Patch-and-Text Prediction (PTP) objective, which masks and recovers both image patches of screenshots and text within screenshots. We also conduct extensive ablation studies on masking rates and patch sizes, as well as designs for improving training stability. Our pre-trained model, while solely taking visual inputs, achieves comparable performance with BERT on 6 out of 8 GLUE tasks (within 2%) and improves up to 8% over prior work. Additionally, we extend PTP to train autoregressive screenshot LMs and demonstrate its effectiveness--our models can significantly reduce perplexity by utilizing the screenshot context. Together, we hope our findings can inspire future research on developing powerful screenshot LMs and extending their reach to broader applications.

For agents

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

get_harvested_code_for_paper("2402.14073")
get_code_for_paper("2402.14073")
have("2402.14073")

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

curl https://syntology.ai/api/ran/2402.14073.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.14073.svg)](https://syntology.ai/paper/2402.14073)

Connect an agent — have() is free.