SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.07544 · CVPR · 2024

Situational Awareness Matters in 3D Vision Language Reasoning

Yu-Xiong Wang, Liang-Yan Gui, Yunze Man

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 7 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/2406.07544.json · badge: /api/ran/2406.07544.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
YunzeMan/Situation3D canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
batch_rotation_vector_to_matrix Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/models/sqa_module.py
code served (permissive licence) · get_code("f2988c8d779e42f8")
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 3, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# batch_rotation_vector_to_matrix expects:
# batch_rot_vec: a tensor of shape (batch_size, 3) containing rotation vectors

batch_size = 2
batch_rot_vec = torch.randn(batch_size, 3)

ARGS = {
    "batch_rot_vec": batch_rot_vec
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 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
decode_eta Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/utils/eta.py
code served (permissive licence) · get_code("86a87fc28c57c91f")
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 dict. 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

# Set seed for reproducibility
torch.manual_seed(0)

# The function takes a single parameter: eta_sec
# It expects a numeric value (int or float) representing seconds
# The function converts this to a dictionary with hours, minutes, seconds

# Test with different values to cover different branches
eta_sec_value = 3665  # 1 hour, 1 minute, 5 seconds

ARGS = {
    'eta_sec': eta_sec_value
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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_eta Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/utils/eta.py
code served (permissive licence) · get_code("e6aaf158a822877c")
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 dict. 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 time

# Construct arguments for get_eta function
# The function calculates estimated time of arrival (ETA) based on:
# - start: start time (float, seconds)
# - end: end time (float, seconds)
# - extra: extra overhead time (float, seconds)
# - num_left: number of items left to process (int)

start = time.time()
end = start + 2.5  # Simulate 2.5 seconds of execution
extra = 1.0  # 1 second of extra overhead per iteration
num_left = 10  # 10 items remaining to process

ARGS = {
    'start': start,
    'end': end,
    'extra': extra,
    'num_left': num_left
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
poly_area Ran
ran, violated the contract we drafted
YunzeMan/Situation3D/situation3d/utils/box_util.py
code served (permissive licence) · get_code("e5929ca970405ef7")
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 float64, ndim 0, 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 numpy as np

# poly_area computes the area of a polygon given x and y coordinates
# It uses np.dot and np.roll, expecting array-like inputs
# x and y should be 1D arrays of coordinates

x = np.array([0.0, 1.0, 1.0, 0.0])
y = np.array([0.0, 0.0, 1.0, 1.0])

ARGS = {
    "x": x,
    "y": y,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
poly_area_batch Ran
ran, honoured the contract we drafted
YunzeMan/Situation3D/situation3d/utils/box_util.py
code served (permissive licence) · get_code("4aa9e3769f269306")
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 1, dtype family float, shape (2,) 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

# poly_area_batch computes polygon areas using the shoelace formula
# It takes x and y coordinate arrays and applies vectorized operations
# x and y should be numpy arrays where each row represents a polygon's coordinates

# Create sample polygon coordinates
# For batch_size=2, num_vertices=4
x = np.array([
    [0.0, 1.0, 1.0, 0.0],
    [0.0, 2.0, 2.0, 0.0]
], dtype=np.float32)

y = np.array([
    [0.0, 0.0, 1.0, 1.0],
    [0.0, 0.0, 2.0, 2.0]
], dtype=np.float32)

ARGS = {
    "x": x,
    "y": y
}
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 1, dtype family float, shape (2,) — shape (2,) 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
polygon_clip Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/utils/box_util.py
code served (permissive licence) · get_code("05b20b55c5918c4a")
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.

# Define a simple polygon - a square in counter-clockwise order
subjectPolygon = [(0, 0), (4, 0), (4, 4), (0, 4)]

# Define a clipping polygon - a convex shape (triangle) in counter-clockwise order
clipPolygon = [(1, 1), (5, 1), (2, 5)]

ARGS = {
    "subjectPolygon": subjectPolygon,
    "clipPolygon": clipPolygon,
}
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
quaternions_to_rotation_matrices Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/models/sqa_module.py
code served (permissive licence) · get_code("f1fe1fca70db75e3")
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 3, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# Create a batch of 2 quaternions in the format (x, y, z, w)
# Quaternions should be normalized but we'll use small values for testing
quaternions = torch.tensor([
    [0.0, 0.0, 0.0, 1.0],  # identity quaternion
    [0.5, 0.5, 0.5, 0.5]   # another quaternion
], dtype=torch.float32)

ARGS = {
    "quaternions": quaternions
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 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
setup_logger Ran
ran; our contract draft was wrong, not the code
YunzeMan/Situation3D/situation3d/utils/logger.py
code served (permissive licence) · get_code("25e18849d43559c1")
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 Logger. 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 functools
import logging
import os
import sys
import tempfile
from termcolor import colored

ARGS = {
    'output': None,
    'distributed_rank': 0,
    'color': True,
    'name': 'test_logger',
    'abbrev_name': 'tl',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
termcolor — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger 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
sqa3d_question_type Ran
ran, honoured the contract we drafted
YunzeMan/Situation3D/situation3d/utils/language_util.py
code served (permissive licence) · get_code("d474ca8ac6d98398")
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 int, ndim 0, dtype family int, 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

# Construct a valid argument for the sqa3d_question_type function
# The function expects a 'record' parameter which is a string
# It splits the string, gets the first word in lowercase, and returns a category number

record = "What is the color of the ball?"

ARGS = {
    "record": record
}
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 int, ndim 0, dtype family int, 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
voc_ap Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
YunzeMan/Situation3D/situation3d/utils/eval_det.py
pointer only (licence: MIT); the stored copy get_code serves, from SupreethN/Hand-CNN, is pointer only (licence: NOASSERTION) · get_code("95c31c78be42f723")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from SupreethN/Hand-CNN, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 2 later) and 6 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
voc_ap 27 other papers
23 dated earlier, 3 later, 1 same year, order not determinable
1506.01497 (2015-06)
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
1605.06409 (2016-05)
R-FCN: Object Detection via Region-based Fully Convolutional Networks
1609.03605 (2016-09)
Detecting Text in Natural Image with Connectionist Text Proposal Network
and 24 more not named here
poly_area 3 other papers
3 dated earlier
2105.07647 (2021-05)
FGR: Frustum-Aware Geometric Reasoning for Weakly Supervised 3D Vehicle Detection
2210.01917 (2022-10)
Differentiable Raycasting for Self-supervised Occupancy Forecasting
2404.16493 (2024-04)
Commonsense Prototype for Outdoor Unsupervised 3D Object Detection
polygon_clip 3 other papers
3 dated earlier
2105.07647 (2021-05)
FGR: Frustum-Aware Geometric Reasoning for Weakly Supervised 3D Vehicle Detection
2210.01917 (2022-10)
Differentiable Raycasting for Self-supervised Occupancy Forecasting
2404.16493 (2024-04)
Commonsense Prototype for Outdoor Unsupervised 3D Object Detection
setup_logger 1 other paper
1 later
2407.09829 (2024-07)
VLMPC: Vision-Language Model Predictive Control for Robotic Manipulation

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

Being able to carry out complicated vision language reasoning tasks in 3D space represents a significant milestone in developing household robots and human-centered embodied AI. In this work, we demonstrate that a critical and distinct challenge in 3D vision language reasoning is the situational awareness, which incorporates two key components: (1) The autonomous agent grounds its self-location based on a language prompt. ( 2 ) The agent answers open-ended questions from the perspective of its calculated position. To address this challenge, we introduce SIG3D, an end-to-end Situation-Grounded model for 3D vision language reasoning. We tokenize the 3D scene into sparse voxel representation, and propose a languagegrounded situation estimator, followed by a situated question answering module. Experiments on the SQA3D and ScanQA datasets show that SIG3D outperforms state-ofthe-art models in situational estimation and question answering by a large margin (e.g., an enhancement of over 30% on situation accuracy). Subsequent analysis corroborates our architectural design choices, explores the distinct functions of visual and textual tokens, and highlights the importance of situational awareness in the domain of 3D question-answering. Project page is available at https: //yunzeman.github.io/situation3d.

For agents

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

get_harvested_code_for_paper("2406.07544")
get_code_for_paper("2406.07544")
have("2406.07544")

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

curl https://syntology.ai/api/ran/2406.07544.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.07544.svg)](https://syntology.ai/paper/2406.07544)

Connect an agent — have() is free.