SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.01852 · ICLR · 2024

LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment

Wei Liu, Zhifeng Li, Li Yuan, Hongfa Wang, Bin Zhu, Bin Lin, Munan Ning, Yang Yan, Jiaxi Cui, Yatian Pang, Wenhao Jiang, Junwu Zhang, and 2 more

arXiv · PDF · Open in the Atlas

Code that ran

10 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; 5 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 4 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/2310.01852.json · badge: /api/ran/2310.01852.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
PKU-YuanGroup/LanguageBind canonical (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
zhihaozhang97/ru-ai community (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
cast_if_src_dtype Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhihaozhang97/ru-ai/imagebind/models/helpers.py
pointer only (licence: MIT); the stored copy get_code serves, from facebookresearch/imagebind, is pointer only (licence: NOASSERTION) · get_code("751b1c8ccbe160c4")
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 facebookresearch/imagebind, 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.

combine_parts Ran
ran on a synthesized input
zhihaozhang97/ru-ai/download_all.py
code served (permissive licence) · get_code("81763ce5edd977e6")
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 str. 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 tempfile
import os

# Create temporary directory and files for testing
temp_dir = tempfile.mkdtemp()

# Create some part files with content
part_files = []
for i in range(3):
    part_file = os.path.join(temp_dir, f"test_part_{i}")
    with open(part_file, 'wb') as f:
        f.write(f"Part {i} content\n".encode())
    part_files.append(part_file)

# Define the base name for the combined file
combined_file = os.path.join(temp_dir, "combined.bin")

ARGS = {
    'base_name': combined_file,
    'parts': part_files
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_clip_timepoints Ran
ran, honoured the contract we drafted
zhihaozhang97/ru-ai/imagebind/data.py
code served (permissive licence) · get_code("432c312da80537d6")
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.

import torch
import numpy as np

# Stand-in clip sampler class that mimics the behavior of a clip sampler
class MockClipSampler:
    def __init__(self, num_clips=3):
        self.num_clips = num_clips
        self.call_count = 0
    
    def __call__(self, end, duration, annotation=None):
        # Returns (start, end, _, _, is_last_clip)
        # The function calls this repeatedly until is_last_clip is True
        clip_duration = duration / self.num_clips
        start = self.call_count * clip_duration
        end = start + clip_duration
        is_last_clip = (self.call_count == self.num_clips - 1)
        
        self.call_count += 1
        
        # Return 5 values: start, end, and three underscores (unused)
        return start, end, None, None, is_last_clip

# Create the clip sampler
clip_sampler = MockClipSampler(num_clips=3)

# Duration in seconds
duration = 10.0

ARGS = {
    "clip_sampler": clip_sampler,
    "duration": duration
}
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 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
get_depth_transform Ran
ran; our contract draft was wrong, not the code
PKU-YuanGroup/LanguageBind/languagebind/depth/processing_depth.py
code served (permissive licence) · get_code("ac6a026c5d4ff47b")
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
from torch import nn
import numpy as np

torch.manual_seed(0)

class MockVisionConfig:
    def __init__(self):
        self.max_depth = 10.0

class MockConfig:
    def __init__(self):
        self.vision_config = MockVisionConfig()

config = MockConfig()

ARGS = {
    "config": config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
get_sinusoid_encoding_table Ran
ran, honoured the contract we drafted
zhihaozhang97/ru-ai/imagebind/models/multimodal_preprocessors.py
pointer only (licence: MIT); the stored copy get_code serves, from howardzh01/ppma, is pointer only (licence: NOASSERTION) · get_code("35a743e50e17e66d")
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 howardzh01/ppma, 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.

int16_to_float32_torch Ran
ran, honoured the contract we drafted
PKU-YuanGroup/LanguageBind/languagebind/audio/processing_audio.py
code served (permissive licence) · get_code("20a58a8c04ccda43")
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 2, dtype family float, shape (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

torch.manual_seed(0)

# The function takes an int16 tensor and converts it to float32
# by dividing by 32767.0 and casting to float32
x = torch.tensor([[-32767, -16384, 0, 16384], [32767, 1000, -1000, 100]], dtype=torch.int16)

ARGS = {
    "x": x
}
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 2, dtype family float, shape (2, 4) — shape (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: none by the harness; any seed is in the setup block
load_and_transform_vision_data Ran
ran on a synthesized input
zhihaozhang97/ru-ai/imagebind/data.py
code served (permissive licence) · get_code("3b8a9225053af887")
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 4, dtype family float, shape (2, 3, 224, 224) 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
from PIL import Image
import tempfile
import os

torch.manual_seed(0)

# Create temporary directory and dummy images
temp_dir = tempfile.mkdtemp()
image_paths = []

for i in range(2):
    # Create a simple RGB image (224x224 to avoid resizing artifacts)
    img = Image.new('RGB', (224, 224), color=(73, 109, 137))
    img_path = os.path.join(temp_dir, f'image_{i}.png')
    img.save(img_path)
    image_paths.append(img_path)

ARGS = {
    'image_paths': image_paths,
    'device': 'cpu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — 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, 224, 224) — shape (2, 3, 224, 224) 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
make_list_of_images Ran
ran, honoured the contract we drafted
PKU-YuanGroup/LanguageBind/languagebind/audio/processing_audio.py
pointer only (licence: MIT); the stored copy get_code serves, from jacklishufan/omniflows, is pointer only (licence: NONE) · get_code("852e87ab903330a7")
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 jacklishufan/omniflows, 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.

opencv_loader Ran
ran, honoured the contract we drafted
PKU-YuanGroup/LanguageBind/languagebind/depth/processing_depth.py
code served (permissive licence) · get_code("5530423121ac4757")
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 3, dtype family float, shape (4, 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 cv2
import numpy as np
import tempfile
import os

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a simple test image (4x3 BGR image)
test_image = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
cv2.imwrite(temp_image_path, test_image)

ARGS = {
    "path": temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 3, 3) — shape (4, 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
stack_dict Ran
ran; our contract draft was wrong, not the code
PKU-YuanGroup/LanguageBind/model/languagebind.py
code served (permissive licence) · get_code("833effea5806a36e")
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

torch.manual_seed(0)

# stack_dict takes:
# 1. x: a list of dicts, where each dict has tensors as values
# 2. device: a torch device

# Create a list of dicts with tensors
x = [
    {
        "tensor_a": torch.randn(2, 4),
        "tensor_b": torch.randn(3, 5),
    },
    {
        "tensor_a": torch.randn(2, 4),
        "tensor_b": torch.randn(3, 5),
    },
]

device = torch.device("cpu")

ARGS = {
    "x": x,
    "device": device,
}
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_deposition_files Not yet run
unverified (harvested, no recorded run)
zhihaozhang97/ru-ai/download_all.py
code served (permissive licence) · get_code("67710298891367de")
How it ran

no invocation recorded.

imagebind_huge Not yet run
unverified (harvested, no recorded run)
zhihaozhang97/ru-ai/imagebind/models/imagebind_model.py
code served (permissive licence) · get_code("bee2c9fa238a8b73")
How it ran

no invocation recorded.

interpolate_pos_encoding Not yet run
unverified (harvested, no recorded run)
zhihaozhang97/ru-ai/imagebind/models/multimodal_preprocessors.py
code served (permissive licence) · get_code("71227efe6a7df1d2")
How it ran

no invocation recorded.

interpolate_pos_encoding_2d Not yet run
unverified (harvested, no recorded run)
zhihaozhang97/ru-ai/imagebind/models/multimodal_preprocessors.py
code served (permissive licence) · get_code("ddd2e52e6b0956b5")
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, 10 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 10 later) and 4 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_sinusoid_encoding_table 21 other papers
10 dated earlier, 11 later · 1 only as an assembly brick
2111.06377 (2021-11)
Masked Autoencoders Are Scalable Vision Learners
2208.07049 (2022-08)
Self-Supervised Vision Transformers for Malware Detection
2302.14042 (2023-02)
Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Images
and 18 more not named here
get_clip_timepoints 6 other papers
2 dated earlier, 4 later
2305.16355 (2023-05)
PandaGPT: One Model To Instruction-Follow Them All
2309.00615 (2023-09)
Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following
2401.10727 (2024-01)
MLLM-Tool: A Multimodal Large Language Model For Tool Agent Learning
and 3 more not named here
cast_if_src_dtype 5 other papers
3 dated earlier, 2 later
2305.05665 (2023-05)
ImageBind: One Embedding Space To Bind Them All
2305.16355 (2023-05)
PandaGPT: One Model To Instruction-Follow Them All
2309.00615 (2023-09)
Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following
and 2 more not named here
interpolate_pos_encoding 5 other papers
3 dated earlier, 2 later
2305.05665 (2023-05)
ImageBind: One Embedding Space To Bind Them All
2305.16355 (2023-05)
PandaGPT: One Model To Instruction-Follow Them All
2309.00615 (2023-09)
Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following
and 2 more not named here
interpolate_pos_encoding_2d 5 other papers
3 dated earlier, 2 later
2305.05665 (2023-05)
ImageBind: One Embedding Space To Bind Them All
2305.16355 (2023-05)
PandaGPT: One Model To Instruction-Follow Them All
2309.00615 (2023-09)
Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following
and 2 more not named here
int16_to_float32_torch 4 other papers
4 later
2311.10122 (2023-11)
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
2407.18062 (2024-07)
Audio Entailment: Assessing Deductive Reasoning for Audio Understanding
2410.02056 (2024-10)
Synthio: Augmenting Small-Scale Audio Classification Datasets with Synthetic Data
and 1 more not named here
make_list_of_images 2 other papers
2 later
2311.10122 (2023-11)
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
2412.01169 (2024-12)
OmniFlow: Any-to-Any Generation with Multi-Modal Rectified Flows
get_depth_transform 1 other paper
1 later
2311.10122 (2023-11)
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
opencv_loader 1 other paper
1 later
2311.10122 (2023-11)
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
stack_dict 1 other paper
1 later
2311.10122 (2023-11)
Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

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

The video-language (VL) pretraining has achieved remarkable improvement in multiple downstream tasks. However, the current VL pretraining framework is hard to extend to multiple modalities (N modalities, N ≥ 3) beyond vision and language. We thus propose LanguageBind, taking the language as the bind across different modalities because the language modality is well-explored and contains rich semantics. Specifically, we freeze the language encoder acquired by VL pretraining and then train encoders for other modalities with contrastive learning. As a result, all modalities are mapped to a shared feature space, implementing multi-modal semantic alignment. While LanguageBind ensures that we can extend VL modalities to N modalities, we also need a high-quality dataset with alignment data pairs centered on language. We thus propose VIDAL-10M with 10 Million data with Video, Infrared, Depth, Audio and their corresponding Language. In our VIDAL-10M, all videos are from short video platforms with complete semantics rather than truncated segments from long videos, and all the video, depth, infrared, and audio modalities are aligned to their textual descriptions. LanguageBind has achieved superior performance on a wide range of 15 benchmarks covering video, audio, depth, and infrared. Moreover, multiple experiments have provided evidence for the effectiveness of LanguageBind in achieving indirect alignment and complementarity among diverse modalities. Code address: https://github.com/PKU-YuanGroup/LanguageBind

For agents

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

get_harvested_code_for_paper("2310.01852")
get_code_for_paper("2310.01852")
have("2310.01852")

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

curl https://syntology.ai/api/ran/2310.01852.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.01852.svg)](https://syntology.ai/paper/2310.01852)

Connect an agent — have() is free.