SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 05527 · ECCV · 2024

A Semantic Space is Worth 256 Language Descriptions: Make Stronger Segmentation Models with Descriptive Properties

Alan Yuille, Zhiding Yu, Jieru Mei, Cihang Xie, Bingchen Zhao, Yuyin Zhou, Junfei Xiao, Ziqi Zhou, Wenxuan Li, Shiyi Lan

Open in the Atlas

Code that ran

4 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; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 1 failure of our instrument; 2 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/05527.json · badge: /api/ran/05527.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
lambert-x/ProLab canonical (classifier label)
classifier basis: census
4 ran · 2 no recorded run
FunctionStatusWhere it lives
get_num_layer_for_vit Ran
ran, honoured the contract we drafted
lambert-x/ProLab/mmcv_custom/layer_decay_optimizer_constructor.py
code served (permissive licence) · get_code("7e302eeb245b5b47")
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

torch.manual_seed(0)

ARGS = {
    'var_name': 'backbone.blocks.5.attn.qkv.weight',
    'num_max_layer': 12
}
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
sa_flops Ran
ran, honoured the contract we drafted
lambert-x/ProLab/get_flops.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from gls0425/linvt, is pointer only (licence: NONE) · get_code("2dcfa48e398dddca")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from gls0425/linvt, and GitHub detected no licence in that repository when it was checked. 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.

weights_to_cpu Ran
ran; our contract draft was wrong, not the code
lambert-x/ProLab/mmcv_custom/my_checkpoint.py
code served (permissive licence) · get_code("b683a315ccc25715")
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 OrderedDict. 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 collections import OrderedDict

torch.manual_seed(0)

# Create a small state dict with GPU tensors (simulated on CPU)
# The function expects an OrderedDict with tensor values
state_dict = OrderedDict()
state_dict['layer1.weight'] = torch.randn(4, 3)
state_dict['layer1.bias'] = torch.randn(2)
state_dict['layer2.weight'] = torch.randn(5, 4)

ARGS = {
    'state_dict': state_dict
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
window_sa_flops Ran
ran on a synthesized input
lambert-x/ProLab/get_flops.py
code served (permissive licence) · get_code("53a7378b123de12f")
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
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# def window_sa_flops(h, w, dim, window_size):
#     return 2 * h * w * window_size * window_size * dim
#
# Parameters:
# - h: height (integer)
# - w: width (integer)
# - dim: dimension (integer)
# - window_size: window size (integer)
#
# All parameters are used in simple arithmetic multiplication,
# so they should be integers or numeric values.

ARGS = {
    "h": 8,
    "w": 12,
    "dim": 4,
    "window_size": 3
}
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 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
get_flops Not yet run
unverified (harvested, no recorded run)
lambert-x/ProLab/get_flops.py
code served (permissive licence) · get_code("8888cf62bd470b9e")
How it ran

no invocation recorded.

my_load_checkpoint Not yet run
unverified (harvested, no recorded run)
lambert-x/ProLab/mmcv_custom/my_checkpoint.py
code served (permissive licence) · get_code("dd4e4ecc5201c127")
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, 3 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 3 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
get_num_layer_for_vit 1 other paper
1 same year, order not determinable
2403.02308 (2024-03)
Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architectures
sa_flops 1 other paper
1 same year, order not determinable
2412.05185 (2024-12)
LinVT: Empower Your Image-level Large Language Model to Understand Videos
weights_to_cpu 1 other paper
1 dated earlier
2308.02097 (2023-08)
Multi-interactive Feature Learning and a Full-time Multi-modality Benchmark for Image Fusion and Segmentation

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 introduce ProLab, a novel approach using propertylevel label space for creating strong interpretable segmentation models. Instead of relying solely on category-specific annotations, ProLab uses descriptive properties grounded in common sense knowledge for supervising segmentation models. It is based on two core designs. First, we employ Large Language Models (LLMs) and carefully crafted prompts to generate descriptions of all involved categories that carry meaningful common sense knowledge and follow a structured format. Second, we introduce a description embedding model preserving semantic correlation across descriptions and then cluster them into a set of descriptive properties (e.g., 256) using K-Means. These properties are based on interpretable common sense knowledge consistent with theories of human recognition. We empirically show that our approach makes segmentation models perform stronger on five classic benchmarks (e.g., ADE20K, COCO-Stuff, Pascal Context, Cityscapes and BDD). Our method also shows better scalability with extended training steps than category-level supervision. Our interpretable segmentation framework also emerges with the generalization ability to segment out-of-domain or unknown categories using in-domain descriptive properties. Code is available at https: //github.com/lambert-x/ProLab.

For agents

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

get_harvested_code_for_paper("05527")
get_code_for_paper("05527")
have("05527")

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

curl https://syntology.ai/api/ran/05527.json

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

[![Syntology run record](https://syntology.ai/api/ran/05527.svg)](https://syntology.ai/paper/05527)

Connect an agent — have() is free.