SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2009.08348 · 2020

S2SD: Simultaneous Similarity-based Self-Distillation for Deep Metric Learning

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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 2 have no contract verdict; 0 failures of our instrument; 1 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/2009.08348.json · badge: /api/ran/2009.08348.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
MLforHealth/S2SD canonical (classifier label)
classifier basis: census
2 ran · 1 no recorded run
FunctionStatusWhere it lives
s2sd_parameters Ran
ran on a synthesized input
MLforHealth/S2SD/parameters.py
code served (permissive licence) · get_code("16a51e8478d5ab48")
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 MockParser. 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 argparse

torch.manual_seed(0)

class MockParser:
    """Stand-in for argparse.ArgumentParser that captures arguments."""
    def __init__(self):
        self.arguments = {}
    
    def add_argument(self, name, **kwargs):
        self.arguments[name] = kwargs
        return self

# Create a parser instance
parser = MockParser()

# The function s2sd_parameters adds arguments to a parser
# We need to call it and capture what arguments it defines
def s2sd_parameters(parser):
    parser.add_argument('--loss_s2sd_source', default='multisimilarity', type=str, help='DML criterion for the base embedding branch.')
    parser.add_argument('--loss_s2sd_target', default='multisimilarity', type=str, help='DML criterion for the target embedding branches.')
    parser.add_argument('--loss_s2sd_T', default=1, type=float, help='Temperature for the KL-Divergence Distillation.')
    parser.add_argument('--loss_s2sd_w', default=50, type=float, help='Weight of the distillation loss.')
    parser.add_argument('--loss_s2sd_pool_aggr', action='store_true', help='Flag. If set, uses both global max- and average pooling in the target branches.')
    parser.add_argument('--loss_s2sd_target_dims', default=[512, 1024, 1536, 2048], nargs='+', type=int, help='Defines number and dimensionality of used target branches.')
    parser.add_argument('--loss_s2sd_feat_distill', action='store_true', help='Flag. If set, feature distillation is used.')
    parser.add_argument('--loss_s2sd_feat_w', default=50, type=float, help='Weight of the feature space distillation loss.')
    parser.add_argument('--loss_s2sd_feat_distill_delay', default=1000, type=int, help='Defines the number of training iterations before feature distillation is activated.')
    return parser

result = s2sd_parameters(parser)

ARGS = {
    'parser': result
}
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 MockParser — a MockParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockParser 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
wandb_parameters Ran
ran on a synthesized input
MLforHealth/S2SD/parameters.py
code served (permissive licence) · get_code("cf110ed7999166a5")
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 ArgumentParser. 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 argparse

# The function takes a parser object and adds arguments to it
# We need to create an ArgumentParser instance to pass to wandb_parameters

parser = argparse.ArgumentParser()

ARGS = {
    'parser': parser
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
basic_training_parameters Not yet run
unverified (harvested, no recorded run)
MLforHealth/S2SD/parameters.py
code served (permissive licence) · get_code("1edf35955d77e643")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 2 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
basic_training_parameters 1 other paper
1 dated earlier
2002.08473 (2020-02)
Revisiting Training Strategies and Generalization Performance in Deep Metric Learning

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

Deep Metric Learning (DML) provides a crucial tool for visual similarity and zero-shot applications by learning generalizing embedding spaces, although recent work in DML has shown strong performance saturation across training objectives. However, generalization capacity is known to scale with the embedding space dimensionality. Unfortunately, high dimensional embeddings also create higher retrieval cost for downstream applications. To remedy this, we propose \emph{Simultaneous Similarity-based Self-distillation (S2SD). S2SD extends DML with knowledge distillation from auxiliary, high-dimensional embedding and feature spaces to leverage complementary context during training while retaining test-time cost and with negligible changes to the training time. Experiments and ablations across different objectives and standard benchmarks show S2SD offers notable improvements of up to 7% in Recall@1, while also setting a new state-of-the-art. Code available at https://github.com/MLforHealth/S2SD.

For agents

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

get_harvested_code_for_paper("2009.08348")
get_code_for_paper("2009.08348")
have("2009.08348")

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

curl https://syntology.ai/api/ran/2009.08348.json

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

[![Syntology run record](https://syntology.ai/api/ran/2009.08348.svg)](https://syntology.ai/paper/2009.08348)

Connect an agent — have() is free.