SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.10864 · NeurIPS · 2022

Cluster and Aggregate: Face Recognition with Large Probe Set

Feng Liu, Xiaoming Liu, Minchul Kim, Anil Jain, Unconstrained Setting

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 1 failure of our instrument; 2 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/2210.10864.json · badge: /api/ran/2210.10864.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
mk-minchul/caface canonical (classifier label)
classifier basis: classifier_2025
7 ran · 2 no recorded run
FunctionStatusWhere it lives
AssignAttention Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("93214fc49b998c82")
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 an instance of its own class, AssignAttention: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "dim": 4,
    "value_dim": 4,
    "num_heads": 2,
    "qkv_bias": False,
    "qk_scale": None,
    "attn_drop": 0.1,
    "proj_drop": 0.1,
    "sum_assign": False,
    "assign_eps": 1.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type AssignAttention — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a AssignAttention 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
AttnBlock Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("e7ef7a4e5e9d4847")
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 an instance of its own class, AttnBlock: the run constructed the object and did not call it, so no result was computed. 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 timm.models.layers import DropPath

torch.manual_seed(0)

ARGS = {
    'dim': 4,
    'num_heads': 2,
    'mlp_ratio': 4.0,
    'qkv_bias': False,
    'qk_scale': None,
    'drop': 0.0,
    'attn_drop': 0.0,
    'drop_path': 0.0,
    'act_layer': nn.GELU,
    'norm_layer': nn.LayerNorm,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type AttnBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a AttnBlock 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
ClusteringBlock Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("d811092eecf8a08a")
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 an instance of its own class, ClusteringBlock: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

# Define a minimal norm layer (LayerNorm)
norm_layer = nn.LayerNorm

# Set up dimensions
dim = 4
value_dim = 4

# Create small test tensors
batch_size = 2
num_tokens = 3
num_group_tokens = 2

x = torch.randn(batch_size, num_tokens, dim)
group_tokens = torch.randn(batch_size, num_group_tokens, dim)
values = torch.randn(batch_size, num_tokens, value_dim)

ARGS = {
    'dim': dim,
    'value_dim': value_dim,
    'norm_layer': norm_layer,
    'sum_assign': False,
    'assign_eps': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ClusteringBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ClusteringBlock 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
L2Norm Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("1d5a95abf105fecc")
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 an instance of its own class, L2Norm: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

# L2Norm.__init__ parameters: channel, dim, eps
# L2Norm.forward uses: x (tensor), torch.norm with p=2, dim, keepdim=True, eps

# Create a small test tensor
# Shape: (batch=2, features=4, spatial=3) to have distinct dimensions
x = torch.randn(2, 4, 3)

ARGS = {
    'channel': None,
    'dim': -1,
    'eps': 1e-05
}
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 L2Norm — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a L2Norm 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
MixerBlock Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("cd507c714a387854")
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 an instance of its own class, MixerBlock: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "tokens_mlp_dim": 16,
    "channels_mlp_dim": 1024,
    "tokens_hidden_dim": 32,
    "channels_hidden_dim": 1024,
    "gated": False,
}
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 MixerBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a MixerBlock 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
MlpBlock Ran
ran on a synthesized input · constructed an object; no result was computed
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("90214f9e6210c7ae")
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 an instance of its own class, MlpBlock: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "input_dim": 4,
    "mlp_dim": 512,
}
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 MlpBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a MlpBlock 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
incremental_mean Ran
ran; our contract draft was wrong, not the code
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("7907c5f114be35a8")
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 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)

ARGS = {
    "prev_mean": torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]]),
    "prev_count": torch.tensor([[2.0, 2.0, 2.0, 2.0], [2.0, 2.0, 2.0, 2.0]]),
    "new_val": torch.tensor([[0.5, 1.5, 2.5, 3.5], [4.5, 5.5, 6.5, 7.5]]),
    "new_val_weight": 1
}
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 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
ClusterAggregateTransformer Not yet run
unverified (harvested, no recorded run)
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("a24a5576d03a6928")
How it ran

no invocation recorded.

ClusteringLayer Not yet run
unverified (harvested, no recorded run)
mk-minchul/caface/caface/nets/transformer/cluster_transformer.py
code served (permissive licence) · get_code("ea798bcd5579da40")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 7 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
AttnBlock
brick of this paper's assembly: helper
3 other papers
3 later · 1 only as an assembly brick
2303.17056 (2023-03)
Audio-Visual Grouping Network for Sound Localization from Mixtures
2309.05281 (2023-09)
Class-Incremental Grouping Network for Continual Audio-Visual Learning
2311.05152 (2023-11)
Cross-modal Prompts: Adapting Large Pre-trained Models for Audio-Visual Downstream Tasks
MlpBlock
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2105.01601 (2021-05)
MLP-Mixer: An all-MLP Architecture for Vision

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

Feature fusion plays a crucial role in unconstrained face recognition where inputs (probes or galleries) comprise of a set of N low quality images whose individual qualities vary. Advances in attention and recurrent modules have led to feature fusion that can model the relationship among the images in the input set. However, attention mechanisms cannot scale to large N due to their quadratic complexity and recurrent modules suffer from input order sensitivity. We propose a two-stage feature fusion paradigm, Cluster and Aggregate, that can both scale to large N and maintain the ability to perform sequential inference with order invariance. Specifically, Cluster stage is a linear assignment of N inputs to M global cluster centers, and Aggregation stage is a fusion over M clustered features. The clustered features play an integral role when the inputs are sequential as they can serve as a summarization of past features. By leveraging the order-invariance of incremental averaging operation, we design an update rule that achieves batch-order invariance, which guarantees that the contributions of early image in the sequence do not diminish as time steps increase. Experiments on IJB-B and IJB-S benchmark datasets show the superiority of the proposed two-stage paradigm in unconstrained face recognition. Code and pretrained models are available in Link.

For agents

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

get_harvested_code_for_paper("2210.10864")
get_code_for_paper("2210.10864")
have("2210.10864")

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

curl https://syntology.ai/api/ran/2210.10864.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.10864.svg)](https://syntology.ai/paper/2210.10864)

Connect an agent — have() is free.