SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2503.04151 · ICCV · 2025

Robust Multi-View Learning via Representation Fusion of Sample-Level Attention and Alignment of Simulated Perturbation

Masashi Sugiyama, Gang Niu, Jie Xu, Na Zhao, Xiaofeng Zhu

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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 5 have no contract verdict; 1 failure of our instrument; 1 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/2503.04151.json · badge: /api/ran/2503.04151.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
SubmissionsIn/RML community
Papers with Code archive
6 ran · 1 no recorded run
FunctionStatusWhere it lives
EncoderLayer Ran
ran on a synthesized input
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("7d0a47fc145a9d6c")
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 tuple. 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 the classes from the provided code
import math
import torch.nn.functional as F

class FeedForward(nn.Module):
    def __init__(self, d_model, d_ff=2048, dropout=0.2):
        super().__init__()
        self.linear_1 = nn.Linear(d_model, d_ff)
        self.dropout_1 = nn.Dropout(dropout)
        self.linear_2 = nn.Linear(d_ff, d_model)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x = self.dropout_1(F.relu(self.linear_1(x)))
        x = self.dropout_2(self.linear_2(x))
        return x

class MultiHeadAttention(nn.Module):
    def __init__(self, heads, d_model, dropout=0.1):
        super().__init__()
        self.d_model = d_model
        self.d_k = d_model // heads
        self.h = heads
        self.q_linear = nn.Linear(d_model, d_model)
        self.v_linear = nn.Linear(d_model, d_model)
        self.k_linear = nn.Linear(d_model, d_model)
        self.dropout = nn.Dropout(dropout)
        self.out = nn.Linear(d_model, d_model)

    def forward(self, q, k, v):
        bs = q.size(0)
        k = self.k_linear(k).view(bs, -1, self.h, self.d_k)
        q = self.q_linear(q).view(bs, -1, self.h, self.d_k)
        v = self.v_linear(v).view(bs, -1, self.h, self.d_k)
        k = k.transpose(1, 2)
        q = q.transpose(1, 2)
        v = v.transpose(1, 2)
        output_scores, scores = attention(q, k, v, self.d_k, self.dropout)
        concat = output_scores.transpose(1, 2).contiguous().view(bs, -1, self.d_model)
        output = self.out(concat)
        return (output, scores)

class Norm(nn.Module):
    def __init__(self, d_model, eps=1e-06):
        super().__init__()
        self.size = d_model
        self.alpha = nn.Parameter(torch.ones(self.size))
        self.bias = nn.Parameter(torch.zeros(self.size))
        self.eps = eps

    def forward(self, x):
        norm = self.alpha * (x - x.mean(dim=-1, keepdim=True)) / (x.std(dim=-1, keepdim=True) + self.eps) + self.bias
        return norm

def attention(q, k, v, d_k, dropout=None):
    scores = torch.matmul(q, k.transpose(-2, -1)) / math.sqrt(d_k)
    scores = F.softmax(scores, dim=-1)
    if dropout is not None:
        scores = dropout(scores)
    output = torch.matmul(scores, v)
    return (output, scores)

class EncoderLayer(nn.Module):
    def __init__(self, d_model, heads, dropout=0.1):
        super().__init__()
        self.norm_1 = Norm(d_model)
        self.norm_2 = Norm(d_model)
        self.attn = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.ff = FeedForward(d_model, dropout=dropout)
        self.dropout_1 = nn.Dropout(dropout)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x2 = self.norm_1(x)
        output, scores = self.attn(x2, x2, x2)
        x = x + self.dropout_1(output)
        x2 = self.norm_2(x)
        x = x + self.dropout_2(self.ff(x2))
        return (x, scores)

INIT_ARGS = {
    'd_model': 4,
    'heads': 2,
    'dropout': 0.1
}

ARGS = {
    'x': torch.randn(2, 3, 4)
}

if __name__ == '__main__':
    encoder_layer = EncoderLayer(**INIT_ARGS)
    output, scores = encoder_layer(**ARGS)
    print("Output shape:", output.shape)
    print("Scores shape:", scores.shape)
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 tuple — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
MultiHeadAttention Ran
ran on a synthesized input
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("0184227dae2b2a3a")
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 tuple. 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 math
import torch.nn.functional as F

def attention(q, k, v, d_k, dropout=None):
    scores = torch.matmul(q, k.transpose(-2, -1)) / math.sqrt(d_k)
    scores = F.softmax(scores, dim=-1)
    if dropout is not None:
        scores = dropout(scores)
    output = torch.matmul(scores, v)
    return (output, scores)

class MultiHeadAttention(nn.Module):

    def __init__(self, heads, d_model, dropout=0.1):
        super().__init__()
        self.d_model = d_model
        self.d_k = d_model // heads
        self.h = heads
        self.q_linear = nn.Linear(d_model, d_model)
        self.v_linear = nn.Linear(d_model, d_model)
        self.k_linear = nn.Linear(d_model, d_model)
        self.dropout = nn.Dropout(dropout)
        self.out = nn.Linear(d_model, d_model)

    def forward(self, q, k, v):
        bs = q.size(0)
        k = self.k_linear(k).view(bs, -1, self.h, self.d_k)
        q = self.q_linear(q).view(bs, -1, self.h, self.d_k)
        v = self.v_linear(v).view(bs, -1, self.h, self.d_k)
        k = k.transpose(1, 2)
        q = q.transpose(1, 2)
        v = v.transpose(1, 2)
        output_scores, scores = attention(q, k, v, self.d_k, self.dropout)
        concat = output_scores.transpose(1, 2).contiguous().view(bs, -1, self.d_model)
        output = self.out(concat)
        return (output, scores)

INIT_ARGS = {
    'heads': 2,
    'd_model': 8,
    'dropout': 0.1
}

ARGS = {
    'q': torch.randn(2, 5, 8),
    'k': torch.randn(2, 5, 8),
    'v': torch.randn(2, 5, 8)
}
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 tuple — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Norm Ran
ran on a synthesized input
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("a87c34abb9f3bffe")
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 (2, 3, 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

INIT_ARGS = {
    "d_model": 4,
    "eps": 1e-06,
}

ARGS = {
    "x": torch.randn(2, 3, 4),
}

if __name__ == "__main__":
    norm = nn.Module()
    norm.__class__ = type('Norm', (nn.Module,), {
        '__init__': lambda self, d_model, eps=1e-06: (
            nn.Module.__init__(self),
            setattr(self, 'size', d_model),
            setattr(self, 'alpha', nn.Parameter(torch.ones(d_model))),
            setattr(self, 'bias', nn.Parameter(torch.zeros(d_model))),
            setattr(self, 'eps', eps),
        ),
        'forward': lambda self, x: (
            self.alpha * (x - x.mean(dim=-1, keepdim=True)) / (x.std(dim=-1, keepdim=True) + self.eps) + self.bias
        ),
    })
    
    norm_instance = norm.__class__(**INIT_ARGS)
    output = norm_instance(**ARGS)
    print(output.shape)
    print(output)
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 (2, 3, 4) — shape (2, 3, 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)
T_Encoders Ran
ran on a synthesized input
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("0bb4e4af64b8087b")
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 tuple. 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 torch.nn.functional as F
import copy
import math

class Norm(nn.Module):
    def __init__(self, d_model, eps=1e-06):
        super().__init__()
        self.size = d_model
        self.alpha = nn.Parameter(torch.ones(self.size))
        self.bias = nn.Parameter(torch.zeros(self.size))
        self.eps = eps

    def forward(self, x):
        norm = self.alpha * (x - x.mean(dim=-1, keepdim=True)) / (x.std(dim=-1, keepdim=True) + self.eps) + self.bias
        return norm

def attention(q, k, v, d_k, dropout=None):
    scores = torch.matmul(q, k.transpose(-2, -1)) / math.sqrt(d_k)
    scores = F.softmax(scores, dim=-1)
    if dropout is not None:
        scores = dropout(scores)
    output = torch.matmul(scores, v)
    return (output, scores)

class MultiHeadAttention(nn.Module):
    def __init__(self, heads, d_model, dropout=0.1):
        super().__init__()
        self.d_model = d_model
        self.d_k = d_model // heads
        self.h = heads
        self.q_linear = nn.Linear(d_model, d_model)
        self.v_linear = nn.Linear(d_model, d_model)
        self.k_linear = nn.Linear(d_model, d_model)
        self.dropout = nn.Dropout(dropout)
        self.out = nn.Linear(d_model, d_model)

    def forward(self, q, k, v):
        bs = q.size(0)
        k = self.k_linear(k).view(bs, -1, self.h, self.d_k)
        q = self.q_linear(q).view(bs, -1, self.h, self.d_k)
        v = self.v_linear(v).view(bs, -1, self.h, self.d_k)
        k = k.transpose(1, 2)
        q = q.transpose(1, 2)
        v = v.transpose(1, 2)
        output_scores, scores = attention(q, k, v, self.d_k, self.dropout)
        concat = output_scores.transpose(1, 2).contiguous().view(bs, -1, self.d_model)
        output = self.out(concat)
        return (output, scores)

class FeedForward(nn.Module):
    def __init__(self, d_model, d_ff=2048, dropout=0.2):
        super().__init__()
        self.linear_1 = nn.Linear(d_model, d_ff)
        self.dropout_1 = nn.Dropout(dropout)
        self.linear_2 = nn.Linear(d_ff, d_model)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x = self.dropout_1(F.relu(self.linear_1(x)))
        x = self.dropout_2(self.linear_2(x))
        return x

class EncoderLayer(nn.Module):
    def __init__(self, d_model, heads, dropout=0.1):
        super().__init__()
        self.norm_1 = Norm(d_model)
        self.norm_2 = Norm(d_model)
        self.attn = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.ff = FeedForward(d_model, dropout=dropout)
        self.dropout_1 = nn.Dropout(dropout)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x2 = self.norm_1(x)
        output, scores = self.attn(x2, x2, x2)
        x = x + self.dropout_1(output)
        x2 = self.norm_2(x)
        x = x + self.dropout_2(self.ff(x2))
        return (x, scores)

def get_clones(module, N):
    return nn.ModuleList([copy.deepcopy(module) for i in range(N)])

class T_Encoders(nn.Module):
    def __init__(self, d_model, N, heads, dropout):
        super().__init__()
        self.N = N
        self.layers = get_clones(EncoderLayer(d_model, heads, dropout), N)
        self.norm = Norm(d_model)

    def forward(self, src):
        x = src
        for i in range(self.N):
            x, scores = self.layers[i](x)
        return (self.norm(x), scores)

INIT_ARGS = {
    'd_model': 4,
    'N': 2,
    'heads': 2,
    'dropout': 0.1
}

ARGS = {
    'src': torch.randn(2, 3, 4)
}
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 tuple — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
TransformerEncoder Ran
ran on a synthesized input
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("385daea0f2dbd0b0")
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 tuple. 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 copy
import math
import torch.nn.functional as F

class Norm(nn.Module):
    def __init__(self, d_model, eps=1e-06):
        super().__init__()
        self.size = d_model
        self.alpha = nn.Parameter(torch.ones(self.size))
        self.bias = nn.Parameter(torch.zeros(self.size))
        self.eps = eps

    def forward(self, x):
        norm = self.alpha * (x - x.mean(dim=-1, keepdim=True)) / (x.std(dim=-1, keepdim=True) + self.eps) + self.bias
        return norm

def attention(q, k, v, d_k, dropout=None):
    scores = torch.matmul(q, k.transpose(-2, -1)) / math.sqrt(d_k)
    scores = F.softmax(scores, dim=-1)
    if dropout is not None:
        scores = dropout(scores)
    output = torch.matmul(scores, v)
    return (output, scores)

class MultiHeadAttention(nn.Module):
    def __init__(self, heads, d_model, dropout=0.1):
        super().__init__()
        self.d_model = d_model
        self.d_k = d_model // heads
        self.h = heads
        self.q_linear = nn.Linear(d_model, d_model)
        self.v_linear = nn.Linear(d_model, d_model)
        self.k_linear = nn.Linear(d_model, d_model)
        self.dropout = nn.Dropout(dropout)
        self.out = nn.Linear(d_model, d_model)

    def forward(self, q, k, v):
        bs = q.size(0)
        k = self.k_linear(k).view(bs, -1, self.h, self.d_k)
        q = self.q_linear(q).view(bs, -1, self.h, self.d_k)
        v = self.v_linear(v).view(bs, -1, self.h, self.d_k)
        k = k.transpose(1, 2)
        q = q.transpose(1, 2)
        v = v.transpose(1, 2)
        output_scores, scores = attention(q, k, v, self.d_k, self.dropout)
        concat = output_scores.transpose(1, 2).contiguous().view(bs, -1, self.d_model)
        output = self.out(concat)
        return (output, scores)

class FeedForward(nn.Module):
    def __init__(self, d_model, d_ff=2048, dropout=0.2):
        super().__init__()
        self.linear_1 = nn.Linear(d_model, d_ff)
        self.dropout_1 = nn.Dropout(dropout)
        self.linear_2 = nn.Linear(d_ff, d_model)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x = self.dropout_1(F.relu(self.linear_1(x)))
        x = self.dropout_2(self.linear_2(x))
        return x

class EncoderLayer(nn.Module):
    def __init__(self, d_model, heads, dropout=0.1):
        super().__init__()
        self.norm_1 = Norm(d_model)
        self.norm_2 = Norm(d_model)
        self.attn = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.ff = FeedForward(d_model, dropout=dropout)
        self.dropout_1 = nn.Dropout(dropout)
        self.dropout_2 = nn.Dropout(dropout)

    def forward(self, x):
        x2 = self.norm_1(x)
        output, scores = self.attn(x2, x2, x2)
        x = x + self.dropout_1(output)
        x2 = self.norm_2(x)
        x = x + self.dropout_2(self.ff(x2))
        return (x, scores)

def get_clones(module, N):
    return nn.ModuleList([copy.deepcopy(module) for i in range(N)])

class T_Encoders(nn.Module):
    def __init__(self, d_model, N, heads, dropout):
        super().__init__()
        self.N = N
        self.layers = get_clones(EncoderLayer(d_model, heads, dropout), N)
        self.norm = Norm(d_model)

    def forward(self, src):
        x = src
        for i in range(self.N):
            x, scores = self.layers[i](x)
        return (self.norm(x), scores)

class TransformerEncoder(nn.Module):
    def __init__(self, d_model, N, heads, dropout):
        super().__init__()
        self.encoders = T_Encoders(d_model, N, heads, dropout)

    def forward(self, src):
        e_outputs, scores = self.encoders(src)
        return (e_outputs, scores)

INIT_ARGS = {
    'd_model': 4,
    'N': 2,
    'heads': 2,
    'dropout': 0.1
}

ARGS = {
    'src': torch.randn(2, 3, 4)
}
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 tuple — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
attention Ran
ran; our contract draft was wrong, not the code
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("62a5ef486fa1cb33")
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 tuple. 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 math

torch.manual_seed(0)

# Create small tensors for q, k, v
# q: (batch=2, seq_len=3, d_k=4)
# k: (batch=2, seq_len=3, d_k=4)
# v: (batch=2, seq_len=3, d_v=5)
q = torch.randn(2, 3, 4)
k = torch.randn(2, 3, 4)
v = torch.randn(2, 3, 5)

# d_k is the dimension of the key (used for scaling)
d_k = 4

# dropout is optional - create a dropout layer
dropout = nn.Dropout(p=0.1)

ARGS = {
    "q": q,
    "k": k,
    "v": v,
    "d_k": d_k,
    "dropout": dropout
}
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 tuple — a tuple output has no array shape of its own; none is 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
Network Not yet run
unverified (harvested, no recorded run)
SubmissionsIn/RML/RML+NRCH/RML_network.py
code served (permissive licence) · get_code("ad129ddc55a8a426")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 1 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 6 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
Norm
brick of this paper's assembly: helper
2 other papers
2 dated earlier · 1 only as an assembly brick
2207.08625 (2022-07)
Unifying Event Detection and Captioning as Sequence Generation via Pre-Training
2303.07180 (2023-03)
Incomplete Multi-View Multi-Label Learning via Label-Guided Masked Viewand Category-Aware Transformers

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

Recently, multi-view learning (MVL) has garnered significant attention due to its ability to fuse discriminative information from multiple views. However, real-world multi-view datasets are often heterogeneous and imperfect, which usually causes MVL methods designed for specific combinations of views to lack application potential and limits their effectiveness. To address this issue, we propose a novel robust MVL method (namely RML) with simultaneous representation fusion and alignment. Specifically, we introduce a simple yet effective multi-view transformer fusion network where we transform heterogeneous multi-view data into homogeneous word embeddings, and then integrate multiple views by the sample-level attention mechanism to obtain a fused representation. Furthermore, we propose a simulated perturbation based multi-view contrastive learning framework that dynamically generates the noise and unusable perturbations for simulating imperfect data conditions. The simulated noisy and unusable data obtain two distinct fused representations, and we utilize contrastive learning to align them for learning discriminative and robust representations. Our RML is self-supervised and can also be applied for downstream tasks as a regularization. In experiments, we employ it in multi-view unsupervised clustering, noise-label classification, and as a plug-and-play module for cross-modal hashing retrieval. Extensive comparison experiments and ablation studies validate RML's effectiveness. Code is available at https://github.com/SubmissionsIn/RML .

For agents

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

get_harvested_code_for_paper("2503.04151")
get_code_for_paper("2503.04151")
have("2503.04151")

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

curl https://syntology.ai/api/ran/2503.04151.json

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

[![Syntology run record](https://syntology.ai/api/ran/2503.04151.svg)](https://syntology.ai/paper/2503.04151)

Connect an agent — have() is free.