SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.06973 · EMNLP · 2024

RWKV-CLIP: A Robust Vision-Language Representation Learner

Jiankang Deng, Weidong Cai, Kaicheng Yang, Dongnan Liu, Tiancheng Gu, An Xiang, Ziyong Feng

arXiv · PDF · Open in the Atlas

Code that ran

7 of 14 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 0 have no contract verdict; 7 failures of our instrument; 7 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/2406.06973.json · badge: /api/ran/2406.06973.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
deepglint/RWKV-CLIP canonical (classifier label)
classifier basis: census
7 ran · 7 no recorded run
FunctionStatusWhere it lives
basic_clean Ran
ran; our contract draft was wrong, not the code
deepglint/RWKV-CLIP/model/open_clip/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.

drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
deepglint/RWKV-CLIP/model/utils_vision_rwkv/drop.py
code served (permissive licence) · get_code("87577b3ff9d32712")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) 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 batch size 2 and 4 dimensions
x = torch.randn(2, 3, 4, 5)

# Define all parameters for drop_path
ARGS = {
    "x": x,
    "drop_prob": 0.5,
    "training": True,
    "scale_by_keep": True,
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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_pairs Ran
ran; our contract draft was wrong, not the code
deepglint/RWKV-CLIP/model/open_clip/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
image_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
deepglint/RWKV-CLIP/model/open_clip/transform.py
code served (permissive licence) · get_code("ff9bfb497bc486af")
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 Compose. 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 torchvision.transforms import InterpolationMode
from dataclasses import dataclass, asdict
from typing import Any, Dict, Optional, Tuple, Union

torch.manual_seed(0)

# Define default constants that the function references
OPENAI_DATASET_MEAN = (0.48145466, 0.4578275, 0.40821073)
OPENAI_DATASET_STD = (0.26862954, 0.26130258, 0.27577711)

# Create minimal AugmentationCfg
@dataclass
class AugmentationCfg:
    scale: Tuple[float, float] = (0.9, 1.0)
    ratio: Optional[Tuple[float, float]] = None
    color_jitter: Optional[Union[float, Tuple[float, float, float], Tuple[float, float, float, float]]] = None
    interpolation: Optional[str] = None
    re_prob: Optional[float] = None
    re_count: Optional[int] = None
    use_timm: bool = False
    color_jitter_prob: float = None
    gray_scale_prob: float = None

# Set up arguments for image_transform function
ARGS = {
    'image_size': 224,
    'is_train': True,
    'mean': (0.48145466, 0.4578275, 0.40821073),
    'std': (0.26862954, 0.26130258, 0.27577711),
    'resize_longest_max': False,
    'fill_color': 0,
    'aug_cfg': AugmentationCfg(
        scale=(0.9, 1.0),
        ratio=None,
        color_jitter=(0.4, 0.4, 0.4, 0.2),
        interpolation=None,
        re_prob=None,
        re_count=None,
        use_timm=False,
        color_jitter_prob=0.8,
        gray_scale_prob=0.2
    )
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
load_model_configs Ran
ran; our contract draft was wrong, not the code
deepglint/RWKV-CLIP/model_config/utils_notebook.py
code served (permissive licence) · get_code("59b9fe12dfcd73e5")
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 Args. 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 json
import tempfile
import os

# Create a temporary config file with the required parameters
config_data = {
    "ctx_len": 1024,
    "head_size": 64,
    "precision": "fp32",
    "input_size": 784,
    "image_patch_size": 14,
    "image_embed_dims": 768,
    "image_num_heads": 12
}

# Create a temporary file and write the config
temp_config = tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=False)
json.dump(config_data, temp_config)
temp_config.close()

ARGS = {
    "config_path": temp_config.name
}
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 Args — a Args output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Args 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
load_model_weight Ran
ran; our contract draft was wrong, not the code
deepglint/RWKV-CLIP/model/utils.py
code served (permissive licence) · get_code("f159f1ed836af624")
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 SimpleModel. 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
import tempfile
import os

torch.manual_seed(0)

# Create a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create model instance
model = SimpleModel()

# Create a state dict and save it to a temporary file
state_dict = model.state_dict()
temp_dir = tempfile.gettempdir()
model_weight_path = os.path.join(temp_dir, "test_model_weight.pt")
torch.save(state_dict, model_weight_path)

# Define ARGS
ARGS = {
    "model": model,
    "model_weight": model_weight_path
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
deepglint/RWKV-CLIP/model/open_clip/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.

WarperCLIP_V_T_RWKV_method Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/model/utils.py
code served (permissive licence) · get_code("4a3350dc35016870")
How it ran

no invocation recorded.

WarperCLIP_V_T_RWKV_text_change_head Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/model/utils.py
code served (permissive licence) · get_code("cd7722094b3b2370")
How it ran

no invocation recorded.

compute_retrieval Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/text_image_retrieval.py
code served (permissive licence) · get_code("4d278aa7e3a2f51e")
How it ran

no invocation recorded.

create_dali_pipeline Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/dataloaders/imagenet.py
code served (permissive licence) · get_code("9da98aebf2dd1b3a")
How it ran

no invocation recorded.

gather_features Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/loss.py
pointer only (licence: MIT); the stored copy get_code serves, from lxa9867/imagefolder, is pointer only (licence: NOASSERTION) · get_code("ddcbd45e940484ee")
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 lxa9867/imagefolder, 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. No invocation recorded.

generate_init_weight_V6 Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/model/Text_rwkv.py
code served (permissive licence) · get_code("f9b03b79fff32f11")
How it ran

no invocation recorded.

get_image_feature Not yet run
unverified (harvested, no recorded run)
deepglint/RWKV-CLIP/text_image_retrieval.py
code served (permissive licence) · get_code("5eaaaa9625688979")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 5 the same month, 5 later) and 9 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
152 dated earlier, 11 the same month, 87 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
133 dated earlier, 9 the same month, 81 later, 7 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
133 dated earlier, 10 the same month, 67 later, 7 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
drop_path 66 other papers
28 dated earlier, 1 the same month, 34 later, 1 same year, order not determinable, 2 undated · 15 only as an assembly brick
1906.01502 (2019-06)
How multilingual is Multilingual BERT?
2011.10566 (2020-11)
Exploring Simple Siamese Representation Learning
2103.03230 (2021-03)
Barlow Twins: Self-Supervised Learning via Redundancy Reduction
and 63 more not named here
gather_features 57 other papers
32 dated earlier, 3 the same month, 21 later, 1 same year, order not determinable · 1 only as an assembly brick
2206.01670 (2022-06)
Egocentric Video-Language Pretraining
2212.00638 (2022-12)
Finetune like you pretrain: Improved finetuning of zero-shot vision models
2301.07094 (2023-01)
Learning Customized Visual Models with Retrieval-Augmented Knowledge
and 54 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

Contrastive Language-Image Pre-training (CLIP) has significantly improved performance in various vision-language tasks by expanding the dataset with image-text pairs obtained from the web. This paper further explores CLIP from the perspectives of data and model architecture. To mitigate the impact of the noise data and enhance the quality of large-scale image-text data crawled from the internet, we introduce a diverse description generation framework that can leverage Large Language Models (LLMs) to combine and refine information from web-based image-text pairs, synthetic captions, and detection tags. Additionally, we propose RWKV-CLIP, the first RWKV-driven vision-language representation learning model that combines the effective parallel training of transformers with the efficient inference of RNNs. Extensive experiments across different model scales and pre-training datasets demonstrate that RWKV-CLIP is a robust vision-language representation learner and it achieves state-of-the-art performance across multiple downstream tasks, including linear probing, zero-shot classification, and zero-shot image-text retrieval. To facilitate future research, the code and pre-trained models are released at https: //github.com/deepglint/RWKV-CLIP.

For agents

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

get_harvested_code_for_paper("2406.06973")
get_code_for_paper("2406.06973")
have("2406.06973")

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

curl https://syntology.ai/api/ran/2406.06973.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.06973.svg)](https://syntology.ai/paper/2406.06973)

Connect an agent — have() is free.