SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.11545 · ICCV · 2023

Bridging Vision and Language Encoders: Parameter-Efficient Tuning for Referring Image Segmentation

Yong Zhang, Yibing Song, Guanbin Li, Xiang Wan, Zunnan Xu, Zhihong Chen

arXiv · PDF · Open in the Atlas

Code that ran

11 of 16 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; 1 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 9 failures of our instrument; 5 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/2307.11545.json · badge: /api/ran/2307.11545.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
kkakkkka/etris canonical (classifier label)
classifier basis: census
10 ran · 4 no recorded run
kkakkkka/ETRIS community
Papers with Code archive
1 ran · 1 no recorded run
FunctionStatusWhere it lives
Interactor Ran
ran on a synthesized input
kkakkkka/ETRIS/model/bridger.py
code served (permissive licence) · get_code("dcafe7ec4c3e2899")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float, shape (3, 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 torch.nn as nn
import torch.nn.functional as F
from torch import Tensor
from typing import List, Optional

def _get_activation_fn(activation):
    """Return an activation function given a string"""
    if activation == 'relu':
        return F.relu
    if activation == 'gelu':
        return F.gelu
    if activation == 'glu':
        return F.glu
    raise RuntimeError(f'activation should be relu/gelu, not {activation}.')

class Interactor(nn.Module):

    def __init__(self, d_model, nhead, dim_feedforward=128, dropout=0.1, activation='relu'):
        super().__init__()
        self.self_attn = nn.MultiheadAttention(d_model, nhead, dropout=dropout)
        self.multihead_attn = nn.MultiheadAttention(d_model, nhead, dropout=dropout)
        self.linear1 = nn.Linear(d_model, dim_feedforward)
        self.dropout = nn.Dropout(dropout)
        self.linear2 = nn.Linear(dim_feedforward, d_model)
        self.norm1 = nn.LayerNorm(d_model)
        self.norm2 = nn.LayerNorm(d_model)
        self.norm3 = nn.LayerNorm(d_model)
        self.dropout1 = nn.Dropout(dropout)
        self.dropout2 = nn.Dropout(dropout)
        self.dropout3 = nn.Dropout(dropout)
        self.activation = _get_activation_fn(activation)

    def with_pos_embed(self, tensor, pos: Optional[Tensor]):
        return tensor if pos is None else tensor + pos

    def forward(self, tgt, memory, tgt_key_padding_mask: Optional[Tensor]=None, memory_key_padding_mask: Optional[Tensor]=None, pos: Optional[Tensor]=None, query_pos: Optional[Tensor]=None):
        q = k = self.with_pos_embed(tgt, query_pos)
        v = tgt
        tgt2 = self.self_attn(q, k, value=v, attn_mask=None, key_padding_mask=tgt_key_padding_mask)[0]
        tgt = tgt + self.dropout1(tgt2)
        tgt = self.norm1(tgt)
        tgt2 = self.multihead_attn(query=self.with_pos_embed(tgt, query_pos), key=self.with_pos_embed(memory, pos), value=memory, attn_mask=None, key_padding_mask=memory_key_padding_mask)[0]
        tgt = tgt + self.dropout2(tgt2)
        tgt = self.norm2(tgt)
        tgt2 = self.linear2(self.dropout(self.activation(self.linear1(tgt))))
        tgt = tgt + self.dropout3(tgt2)
        tgt = self.norm3(tgt)
        return tgt

INIT_ARGS = {
    'd_model': 4,
    'nhead': 2,
    'dim_feedforward': 8,
    'dropout': 0.1,
    'activation': 'relu'
}

ARGS = {
    'tgt': torch.randn(3, 2, 4),
    'memory': torch.randn(5, 2, 4),
    'tgt_key_padding_mask': None,
    'memory_key_padding_mask': None,
    'pos': torch.randn(5, 2, 4),
    'query_pos': torch.randn(3, 2, 4)
}

if __name__ == '__main__':
    model = Interactor(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: {ARGS['tgt'].shape}")
    assert output.shape == ARGS['tgt'].shape, f"Shape mismatch: {output.shape} vs {ARGS['tgt'].shape}"
    print("Success!")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 (3, 2, 4) — shape (3, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
adapt_position_encoding Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/model/clip.py
pointer only (licence: MIT); the stored copy get_code serves, from junbum766/ALRIS, is pointer only (licence: NONE) · get_code("98ab2f374bc768a9")
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 junbum766/ALRIS, 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.

basic_clean Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/utils/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("98f385d847636a3e")
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 ltpwy/MSCI, 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.

conv_layer Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/model/layers.py
pointer only (licence: MIT); the stored copy get_code serves, from franciszzj/tp-sis, is pointer only (licence: NONE) · get_code("dbfe3357dc25ca5f")
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 franciszzj/tp-sis, 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.

deconv_layer Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/model/layers.py
pointer only (licence: MIT); the stored copy get_code serves, from junbum766/ALRIS, is pointer only (licence: NONE) · get_code("db55fb2ff4ac93bb")
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 junbum766/ALRIS, 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.

get_pairs Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/utils/simple_tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
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 set. 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

torch.manual_seed(0)

# The function get_pairs takes a word parameter
# A word is represented as a tuple of symbols (variable-length strings)
# The function iterates through the word and creates pairs of consecutive symbols

word = ("hello", "world", "test")

ARGS = {
    "word": word
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
init_random_seed Ran
ran, honoured the contract we drafted
kkakkkka/etris/utils/misc.py
code served (permissive licence) · get_code("539cbfa1328eb51b")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
import torch
import torch.distributed as dist

torch.manual_seed(0)
np.random.seed(0)

ARGS = {
    'seed': None,
    'device': 'cpu',
    'rank': 0,
    'world_size': 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 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
linear_layer Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/model/layers.py
pointer only (licence: MIT); the stored copy get_code serves, from franciszzj/tp-sis, is pointer only (licence: NONE) · get_code("89f9fd49ddec76ba")
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 franciszzj/tp-sis, 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.

load_cfg_from_cfg_file Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/utils/config.py
pointer only (licence: MIT); the stored copy get_code serves, from arya-domain/UA-VLS, is pointer only (licence: NONE) · get_code("cc23416cff396c39")
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 arya-domain/UA-VLS, 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.

trainMetricGPU Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/utils/misc.py
code served (permissive licence) · get_code("78241a45492d6667")
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, 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

torch.manual_seed(0)

# The function takes output and target tensors that must:
# 1. Have the same shape
# 2. Have dim in [2, 3, 4]
# 3. Be flattened to 2D (batch_size, features)
# 4. Support sigmoid, comparison operations, and bool conversion

# Create small tensors with batch_size=2
# Using dim=4 to satisfy the assertion and have a reasonable flattened size
batch_size = 2
spatial_dims = (3, 2)  # 3x2 spatial, will flatten to 6 features per sample

output = torch.randn(batch_size, 1, 3, 2)  # Shape: (2, 1, 3, 2) - dim=4
target = torch.randint(0, 2, (batch_size, 1, 3, 2)).float()  # Same shape, binary values

ARGS = {
    "output": output,
    "target": target,
    "threshold": 0.35,
    "pr_iou": 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 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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
kkakkkka/etris/utils/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("9542161e9640b858")
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 ltpwy/MSCI, 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.

Bridger_RN Not yet run
unverified (harvested, no recorded run)
kkakkkka/ETRIS/model/bridger.py
code served (permissive licence) · get_code("23af313e4b569833")
How it ran

no invocation recorded.

build_model Not yet run
unverified (harvested, no recorded run)
kkakkkka/etris/model/clip.py
code served (permissive licence) · get_code("a77119da98b6e547")
How it ran

no invocation recorded.

concat_all_gather Not yet run
unverified (harvested, no recorded run)
kkakkkka/etris/utils/misc.py
code served (permissive licence) · get_code("73cecca9f3575f09")
How it ran

no invocation recorded.

loads_pyarrow Not yet run
unverified (harvested, no recorded run)
kkakkkka/etris/utils/dataset.py
code served (permissive licence) · get_code("b4c154b6f0568b6b")
How it ran

no invocation recorded.

merge_cfg_from_list Not yet run
unverified (harvested, no recorded run)
kkakkkka/etris/utils/config.py
pointer only (licence: MIT); the stored copy get_code serves, from arya-domain/UA-VLS, is pointer only (licence: NONE) · get_code("022883c7911679b3")
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 arya-domain/UA-VLS, 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. No invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 13 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 4 the same month, 13 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_pairs 258 other papers
61 dated earlier, 4 the same month, 185 later, 7 same year, order not determinable, 1 undated · 4 only as an assembly brick
1802.06474 (2018-02)
A Closed-form Solution to Photorealistic Image Stylization
1803.01128 (2018-03)
Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examples
1809.00125 (2018-09)
Simple Fusion: Return of the Language Model
and 255 more not named here
whitespace_clean 231 other papers
45 dated earlier, 4 the same month, 176 later, 5 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 228 more not named here
basic_clean 218 other papers
45 dated earlier, 4 the same month, 163 later, 5 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 215 more not named here
concat_all_gather 138 other papers
80 dated earlier, 3 the same month, 54 later, 1 same year, order not determinable · 33 only as an assembly brick
1710.09412 (2017-10)
mixup: BEYOND EMPIRICAL RISK MINIMIZATION
1905.04899 (2019-05)
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
and 135 more not named here
merge_cfg_from_list 23 other papers
12 dated earlier, 10 later, 1 same year, order not determinable
2003.13328 (2020-03)
Strip Pooling: Rethinking Spatial Pooling for Scene Parsing
2108.05293 (2021-08)
Few-Shot Segmentation with Global and Local Contrastive Learning
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
and 20 more not named here
load_cfg_from_cfg_file 22 other papers
11 dated earlier, 10 later, 1 same year, order not determinable
2003.13328 (2020-03)
Strip Pooling: Rethinking Spatial Pooling for Scene Parsing
2108.05293 (2021-08)
Few-Shot Segmentation with Global and Local Contrastive Learning
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
and 19 more not named here
conv_layer 9 other papers
3 dated earlier, 6 later
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
2210.10276 (2022-10)
CLIP-Driven Fine-grained Text-Image Person Re-identification
2306.09244 (2023-06)
Text Promptable Surgical Instrument Segmentation with Vision-Language Models
and 6 more not named here
linear_layer 7 other papers
3 dated earlier, 4 later
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
2210.10276 (2022-10)
CLIP-Driven Fine-grained Text-Image Person Re-identification
2306.09244 (2023-06)
Text Promptable Surgical Instrument Segmentation with Vision-Language Models
and 4 more not named here
loads_pyarrow 6 other papers
5 dated earlier, 1 later
2110.07402 (2021-10)
Self-Supervised Learning by Estimating Twin Class Distributions
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
2202.07800 (2022-02)
Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
and 3 more not named here
adapt_position_encoding 2 other papers
2 later
2501.08580 (2025-01)
Densely Connected Parameter-Efficient Tuning for Referring Image Segmentation
2608.30621 (2026-08)
Cost-efficient Active Learning for Referring Image Segmentation and Grounding
deconv_layer 2 other papers
2 later
2501.08580 (2025-01)
Densely Connected Parameter-Efficient Tuning for Referring Image Segmentation
2608.30621 (2026-08)
Cost-efficient Active Learning for Referring Image Segmentation and Grounding
init_random_seed 2 other papers
1 dated earlier, 1 later
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
2501.08580 (2025-01)
Densely Connected Parameter-Efficient Tuning for Referring Image Segmentation
trainMetricGPU 2 other papers
1 dated earlier, 1 later
2111.15174 (2021-11)
CRIS: CLIP-Driven Referring Image Segmentation
2501.08580 (2025-01)
Densely Connected Parameter-Efficient Tuning for Referring Image 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

Parameter Efficient Tuning (PET) has gained attention for reducing the number of parameters while maintaining performance and providing better hardware resource savings, but few studies investigate dense prediction tasks and interaction between modalities. In this paper, we do an investigation of efficient tuning problems on referring image segmentation. We propose a novel adapter called Bridger to facilitate cross-modal information exchange and inject task-specific information into the pre-trained model. We also design a lightweight decoder for image segmentation. Our approach achieves comparable or superior performance with only 1.61% to 3.38% backbone parameter updates, evaluated on challenging benchmarks. The code is available at https://github.com/kkakkkka/ETRIS.

For agents

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

get_harvested_code_for_paper("2307.11545")
get_code_for_paper("2307.11545")
have("2307.11545")

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

curl https://syntology.ai/api/ran/2307.11545.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.11545.svg)](https://syntology.ai/paper/2307.11545)

Connect an agent — have() is free.