SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.14043 · CVPR · 2022

Semantic Segmentation by Early Region Proxy

Bo Pang, Cewu Lu, Yifan Zhang

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 2 have no contract verdict; 0 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/2203.14043.json · badge: /api/ran/2203.14043.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
YiF-Zhang/RegionProxy canonical (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
FunctionStatusWhere it lives
convert_vit Ran
ran on a synthesized input
YiF-Zhang/RegionProxy/utils/checkpoint.py
code served (permissive licence) · get_code("14b73712f9d05d95")
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)

# Construct a checkpoint dict that resembles a ViT checkpoint
# The function iterates over items and checks key prefixes
ckpt = OrderedDict()

# Add some patch_embed keys (non-hybrid case)
ckpt['patch_embed.proj.weight'] = torch.randn(768, 3, 16, 16)
ckpt['patch_embed.proj.bias'] = torch.randn(768)

# Add norm key
ckpt['norm.weight'] = torch.randn(768)
ckpt['norm.bias'] = torch.randn(768)

# Add blocks with various sublayers
ckpt['blocks.0.norm1.weight'] = torch.randn(768)
ckpt['blocks.0.norm1.bias'] = torch.randn(768)
ckpt['blocks.0.attn.qkv.weight'] = torch.randn(2304, 768)
ckpt['blocks.0.attn.qkv.bias'] = torch.randn(2304)
ckpt['blocks.0.attn.proj.weight'] = torch.randn(768, 768)
ckpt['blocks.0.attn.proj.bias'] = torch.randn(768)
ckpt['blocks.0.norm2.weight'] = torch.randn(768)
ckpt['blocks.0.norm2.bias'] = torch.randn(768)
ckpt['blocks.0.mlp.fc1.weight'] = torch.randn(3072, 768)
ckpt['blocks.0.mlp.fc1.bias'] = torch.randn(3072)
ckpt['blocks.0.mlp.fc2.weight'] = torch.randn(768, 3072)
ckpt['blocks.0.mlp.fc2.bias'] = torch.randn(768)

# Add a head key (which should be skipped)
ckpt['head.weight'] = torch.randn(1000, 768)
ckpt['head.bias'] = torch.randn(1000)

# Add cls_token and pos_embed
ckpt['cls_token'] = torch.randn(1, 1, 768)
ckpt['pos_embed'] = torch.randn(1, 197, 768)

ARGS = {
    'ckpt': ckpt
}
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
old_convert_vit Ran
ran on a synthesized input
YiF-Zhang/RegionProxy/utils/checkpoint.py
code served (permissive licence) · get_code("57d9de80b4fa6fb2")
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
import torch.nn as nn
from collections import OrderedDict

torch.manual_seed(0)

# Create a minimal model with a 'stem' attribute for testing
class MinimalViT(nn.Module):
    def __init__(self):
        super().__init__()
        self.stem = nn.Linear(4, 4)
        self.layers = nn.ModuleList([nn.Linear(4, 4) for _ in range(2)])
    
    def forward(self, x):
        return x

# Create a sample checkpoint with various keys that the function processes
ckpt = OrderedDict()

# Keys that should be skipped (head)
ckpt['head.weight'] = torch.randn(4, 4)
ckpt['head.bias'] = torch.randn(4)

# norm keys -> ln1
ckpt['norm.weight'] = torch.randn(4)
ckpt['norm.bias'] = torch.randn(4)

# patch_embed keys with stem
ckpt['backbone.patch_embed.proj.weight'] = torch.randn(4, 3)
ckpt['backbone.patch_embed.proj.bias'] = torch.randn(4)

# blocks with various sub-keys
ckpt['blocks.0.norm1.weight'] = torch.randn(4)
ckpt['blocks.0.norm1.bias'] = torch.randn(4)
ckpt['blocks.0.norm2.weight'] = torch.randn(4)
ckpt['blocks.0.mlp.fc1.weight'] = torch.randn(4, 4)
ckpt['blocks.0.mlp.fc1.bias'] = torch.randn(4)
ckpt['blocks.0.mlp.fc2.weight'] = torch.randn(4, 4)
ckpt['blocks.0.mlp.fc2.bias'] = torch.randn(4)
ckpt['blocks.0.attn.qkv.weight'] = torch.randn(12, 4)
ckpt['blocks.0.attn.qkv.bias'] = torch.randn(12)
ckpt['blocks.0.attn.proj.weight'] = torch.randn(4, 4)
ckpt['blocks.0.attn.proj.bias'] = torch.randn(4)

# Additional blocks entry
ckpt['blocks.1.norm1.weight'] = torch.randn(4)

# Create model instance
model = MinimalViT()

ARGS = {
    'ckpt': ckpt,
    'model': model
}
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

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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 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.

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

Typical vision backbones manipulate structured features. As a compromise, semantic segmentation has long been modeled as per-point prediction on dense regular grids. In this work, we present a novel and efficient modeling that starts from interpreting the image as a tessellation of learnable regions, each of which has flexible geometrics and carries homogeneous semantics. To model region-wise context, we exploit Transformer to encode regions in a sequence-to-sequence manner by applying multi-layer self-attention on the region embeddings, which serve as proxies of specific regions. Semantic segmentation is now carried out as per-region prediction on top of the encoded region embeddings using a single linear classifier, where a decoder is no longer needed. The proposed RegProxy model discards the common Cartesian feature layout and operates purely at region level. Hence, it exhibits the most competitive performance-efficiency trade-off compared with the conventional dense prediction methods. For example, on ADE20K, the small-sized RegProxy-S/16 outperforms the best CNN model using 25% parameters and 4% computation, while the largest RegProxy-L/16 achieves 52.9 mIoU which outperforms the state-of-the-art by 2.1% with fewer resources. Codes and models are available at https://github.com/YiF-Zhang/RegionProxy.

For agents

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

get_harvested_code_for_paper("2203.14043")
get_code_for_paper("2203.14043")
have("2203.14043")

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

curl https://syntology.ai/api/ran/2203.14043.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.14043.svg)](https://syntology.ai/paper/2203.14043)

Connect an agent — have() is free.