SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.14732 · 2026

DeepMoLM: Leveraging Visual and Geometric Structural Information for Molecule-Text Modeling

Yunlin Mao, Yufeng Jiang, Jung Yoo, Hexiao Ding, Hongzhao Chen, Jing Lan, Gwing Yip, Gerald Cheng, Jing Cai, Liang-Ting Lin, Nga-Chun Ng

arXiv · PDF

Code that ran

14 of 17 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 11 failures 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/2601.14732.json · badge: /api/ran/2601.14732.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
1anj/DeepMoLM canonical (classifier label)
classifier basis: census
14 ran · 3 no recorded run
FunctionStatusWhere it lives
Dynamic_quantize_torch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
1anj/DeepMoLM/llava/model/FloatPointQuantizeTorch.py
code served (permissive licence) · get_code("733a4a5be1673d02")
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)

# Create a small tensor with float values
x = torch.randn(2, 4, dtype=torch.float32)

# bit parameter - number of bits for quantization
bit = 8

# stochastic parameter - whether to use stochastic quantization
stochastic = False

ARGS = {
    'x': x,
    'bit': bit,
    'stochastic': stochastic
}
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
decode_time_token Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/eval/eval_model_loc.py
code served (permissive licence) · get_code("61a0bf7498d3fa37")
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 re

# Construct arguments for decode_time_token function

text = "The event starts at <t0> and ends at <t5> with a checkpoint at <t2>"
duration = 10.0
num_time_tokens = 6
time_token_format = "<t{t}>"

ARGS = {
    "text": text,
    "duration": duration,
    "num_time_tokens": num_time_tokens,
    "time_token_format": time_token_format,
}
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
draw_bounding_boxes Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/eval/model_refcoco.py
code served (permissive licence) · get_code("3890b2d2104caa11")
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 Image. 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.

from PIL import Image, ImageDraw
import torch

# Create a simple test image (RGB, 100x100 pixels)
test_image = Image.new('RGB', (100, 100), color='white')

# Create bounding boxes: list of [x_min, y_min, x_max, y_max]
bboxes = [
    [10, 15, 50, 60],
    [30, 25, 80, 75]
]

# Create labels corresponding to each bounding box
labels = ['cat', 'dog']

# Define all arguments for the function
ARGS = {
    'image': test_image,
    'bboxes': bboxes,
    'labels': labels,
    'color': 'green',
    'width': 2,
    'fontcolor': 'red'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
extract_all_bounding_boxes Ran
ran, honoured the contract we drafted
1anj/DeepMoLM/llava/eval/model_refcoco.py
code served (permissive licence) · get_code("7d4191be32e4ab73")
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.

import re

def extract_all_bounding_boxes(text, w, h):
    pattern = '\\[([\\d.]+),\\s*([\\d.]+),\\s*([\\d.]+),\\s*([\\d.]+)\\]'
    matches = re.findall(pattern, text)
    bboxes = [[float('0.' + b) for b in box] for box in matches]
    bboxes = [[box[0] * w, box[1] * h, box[2] * w, box[3] * h] for box in bboxes]
    return bboxes

ARGS = {
    'text': '[1,2,3,4] some text [5,6,7,8] more text [9,0,1,2]',
    'w': 640,
    'h': 480
}
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
floatExM0_quantize_torch Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/model/FloatPointQuantizeTorch.py
code served (permissive licence) · get_code("556e6e721bc79ca5")
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
import math

torch.manual_seed(0)

# Create a small tensor with positive values (log2 requires positive values)
x = torch.tensor([[1.5, 2.0, 3.5, 4.2],
                   [0.8, 1.2, 2.5, 3.8]], dtype=torch.float32)

e_bit = 4
stochastic = False

ARGS = {
    "x": x,
    "e_bit": e_bit,
    "stochastic": stochastic
}
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
floatExMy_quantize_torch Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/model/FloatPointQuantizeTorch.py
code served (permissive licence) · get_code("a3b94d289d42dbb0")
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)

# Create a small tensor with floating point values
# The function expects a tensor with real numbers to quantize
x = torch.tensor([1.5, 2.3, 0.7, 3.2, 1.1, 2.9, 0.5, 1.8], dtype=torch.float32).reshape(2, 4)

# e_bit: number of exponent bits
e_bit = 4

# m_bit: number of mantissa bits
m_bit = 3

# stochastic: whether to add stochastic noise during quantization
stochastic = False

ARGS = {
    'x': x,
    'e_bit': e_bit,
    'm_bit': m_bit,
    'stochastic': stochastic
}
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
fprints_from_fprints_dict Ran
ran, honoured the contract we drafted
1anj/DeepMoLM/e3fp/pipeline.py
code served (permissive licence) · get_code("9a1ddba46c3665f5")
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 3, 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

# Construct a fprints_dict with multiple levels
# The function calls .get(level, fprints_dict[max(fprints_dict.keys())])
# so we need a dict with integer keys mapping to fingerprint lists

fprints_dict = {
    -1: [torch.randn(2, 4), torch.randn(2, 4)],  # level -1
    0: [torch.randn(2, 4), torch.randn(2, 4)],   # level 0
    1: [torch.randn(2, 4), torch.randn(2, 4)],   # level 1
}

ARGS = {
    "fprints_dict": fprints_dict,
    "level": -1,
}
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 3, 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
get_chunk Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/eval/model_vqa_video.py
code served (permissive licence) · get_code("42a46570620cd9fa")
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 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 math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

def get_chunk(lst, n, k):
    chunks = split_list(lst, n)
    return chunks[k]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3
k = 1

ARGS = {
    "lst": lst,
    "n": n,
    "k": k
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
get_frame_from_vcap Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/mm_utils.py
code served (permissive licence) · get_code("996492ef7b3a48c3")
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 cv2
import numpy as np
from PIL import Image
from unittest.mock import Mock

# Create a mock video capture object
vidcap = Mock()

# Set up mock return values for cv2.CAP_PROP_FPS and cv2.CAP_PROP_FRAME_COUNT
vidcap.get.side_effect = lambda prop: {
    cv2.CAP_PROP_FPS: 30.0,
    cv2.CAP_PROP_FRAME_COUNT: 100.0
}.get(prop, 0.0)

# Create a mock frame (numpy array in BGR format)
mock_frame = np.random.randint(0, 256, (720, 1280, 3), dtype=np.uint8)

# Set up the read method to return success and frame for the first several calls
read_count = [0]
def mock_read():
    read_count[0] += 1
    if read_count[0] <= 100:
        return (True, mock_frame)
    else:
        return (False, None)

vidcap.read.side_effect = mock_read

ARGS = {
    'vidcap': vidcap,
    'num_frames': 10,
    'max_fps': 0.0,
    'fps': 30.0,
    'frame_count': 100,
    'video_file_name': 'test_video.mp4'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy — 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
get_frame_from_vcap_with_fps Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/mm_utils.py
code served (permissive licence) · get_code("7affd35c8ec5c7a7")
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 cv2
import numpy as np
from PIL import Image
import tempfile
import os

# Create a minimal video file for testing
def create_test_video(filename, num_frames=30, fps=10, width=720, height=720):
    """Create a simple test video file"""
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(filename, fourcc, fps, (width, height))
    
    for i in range(num_frames):
        # Create a simple frame with changing color
        frame = np.zeros((height, width, 3), dtype=np.uint8)
        frame[:, :, 0] = (i * 8) % 256  # Blue channel
        frame[:, :, 1] = (i * 4) % 256  # Green channel
        frame[:, :, 2] = (i * 2) % 256  # Red channel
        out.write(frame)
    
    out.release()

# Create a temporary video file
temp_dir = tempfile.gettempdir()
video_path = os.path.join(temp_dir, "test_video_temp.mp4")
create_test_video(video_path, num_frames=30, fps=10, width=720, height=720)

# Open the video with cv2.VideoCapture
vidcap = cv2.VideoCapture(video_path)

ARGS = {
    "vidcap": vidcap,
    "num_frames": 10,
    "max_fps": 5.0,
    "fps": None,
    "frame_count": None,
    "video_file_name": "test_video.mp4"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy — 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
iou Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
1anj/DeepMoLM/llava/eval/eval_model_loc.py
code served (permissive licence) · get_code("5e821d4715fc98c0")
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 float, ndim 0, dtype family float, shape () 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
import numpy as np

# The function takes two tuples of floats representing intervals/segments
# s1[0] is the start of segment 1, s1[1] is the end of segment 1
# s2[0] is the start of segment 2, s2[1] is the end of segment 2
# The function computes Intersection over Union (IoU) of these two intervals

# Create valid arguments: two tuples with (start, end) coordinates
s1 = (1.0, 5.0)  # segment 1: from 1.0 to 5.0
s2 = (3.0, 7.0)  # segment 2: from 3.0 to 7.0

ARGS = {
    "s1": s1,
    "s2": s2
}
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 float, ndim 0, dtype family float, shape () — shape () 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
postprocess_2d_grounding Ran
ran, honoured the contract we drafted
1anj/DeepMoLM/llava/eval/model_refcoco.py
code served (permissive licence) · get_code("16caadc6ca70c092")
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

# Analyze the function to understand what it needs:
# - output: a string that contains coordinates in format like "[x1, y1, x2, y2]" or similar
# - h: height (will be converted to float)
# - w: width (will be converted to float)
# - box_format: string that determines parsing and scaling behavior

# Test case 1: llava format with normalized coordinates (0.xxx)
output_str = "[0.1, 0.2, 0.5, 0.8]."
h = 480
w = 640
box_format = 'llava'

ARGS = {
    'output': output_str,
    'h': h,
    'w': w,
    'box_format': box_format
}
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
precision Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/eval/eval_model_loc.py
code served (permissive licence) · get_code("17bd792b6162fc98")
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 function. 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
from typing import Tuple

def iou(s1: Tuple[float, float], s2: Tuple[float, float]) -> float:
    i = max(min(s1[1], s2[1]) - max(s1[0], s2[0]), 0)
    u = max(s1[1] - s1[0], 0) + max(s2[1] - s2[0], 0) - i
    return i / u if u > 0 else 0

def precision(threshold: float):
    def precision_func(s1: Tuple[float, float], s2: Tuple[float, float]) -> float:
        return float(iou(s1, s2) >= threshold)
    return precision_func

# Setup code
torch.manual_seed(0)

ARGS = {
    "threshold": 0.5
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
split_list Ran
ran; our contract draft was wrong, not the code
1anj/DeepMoLM/llava/eval/model_vqa_video.py
code served (permissive licence) · get_code("076c252c52cbb161")
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 list. 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 math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3

ARGS = {
    "lst": lst,
    "n": n
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 — a list 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
confs_from_smiles Not yet run
unverified (harvested, no recorded run)
1anj/DeepMoLM/e3fp/pipeline.py
code served (permissive licence) · get_code("eefc458e0f161a30")
How it ran

no invocation recorded.

opencv_extract_frames Not yet run
unverified (harvested, no recorded run)
1anj/DeepMoLM/llava/mm_utils.py
code served (permissive licence) · get_code("a8f8416a7383a2fe")
How it ran

no invocation recorded.

params_to_dicts Not yet run
unverified (harvested, no recorded run)
1anj/DeepMoLM/e3fp/pipeline.py
code served (permissive licence) · get_code("8d4494ab7bb19a0f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 2 the same month, 2 later) and 15 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
split_list 197 other papers
186 dated earlier, 2 the same month, 8 later, 1 undated
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 194 more not named here
get_chunk 189 other papers
179 dated earlier, 2 the same month, 8 later
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 186 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

AI models for drug discovery and chemical literature mining must interpret molecular images and generate outputs consistent with 3D geometry and stereochemistry. Most molecular language models rely on strings or graphs, while vision-language models often miss stereochemical details and struggle to map continuous 3D structures into discrete tokens. We propose DeepMoLM: Deep Molecular Language Modeling, a dual-view framework that grounds high-resolution molecular images in geometric invariants derived from molecular conformations. DeepMoLM preserves high-frequency evidence from 1024 × 1024 inputs, encodes conformer neighborhoods as discrete Extended 3-Dimensional Fingerprints, and fuses visual and geometric streams with cross-attention, enabling physically grounded generation without atom coordinates. DeepMoLM improves PubChem captioning with a 12.3% relative METEOR gain over the strongest generalist baseline while staying competitive with specialist methods. It produces valid numeric outputs for all property queries and attains MAE 13.64 g/mol on Molecular Weight and 37.89 on Complexity in the specialist setting. On ChEBI-20 description generation from images, it exceeds generalist baselines and matches state-ofthe-art vision-language models. Code is available at https://github.com/1anj/DeepMoLM.

For agents

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

get_harvested_code_for_paper("2601.14732")
get_code_for_paper("2601.14732")
have("2601.14732")

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

curl https://syntology.ai/api/ran/2601.14732.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.14732.svg)](https://syntology.ai/paper/2601.14732)

Connect an agent — have() is free.