SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1902.03393 · 2019

Improved Knowledge Distillation via Teacher Assistant

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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, 1 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 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/1902.03393.json · badge: /api/ran/1902.03393.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
imirzadeh/Teacher-Assistant-Knowledge-Distillation canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
create_cnn_model Ran
ran; our contract draft was wrong, not the code
imirzadeh/Teacher-Assistant-Knowledge-Distillation/model_factory.py
code served (permissive licence) · get_code("b309f216eed0c475")
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 SimpleResNet. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Stand-in classes for the model books and ConvNetMaker
class SimpleResNet(nn.Module):
    def __init__(self, num_classes=100):
        super().__init__()
        self.fc = nn.Linear(10, num_classes)
    
    def forward(self, x):
        return self.fc(x)

class ConvNetMaker(nn.Module):
    def __init__(self, model_spec):
        super().__init__()
        self.model_spec = model_spec
        self.fc = nn.Linear(10, model_spec.get('num_classes', 100))
    
    def forward(self, x):
        return self.fc(x)

class ModelSpec(dict):
    pass

# Create the model books that the function expects
resnet_book = {
    '110': SimpleResNet,
    '32': SimpleResNet,
}

plane_cifar10_book = {
    '2': ModelSpec({'num_classes': 10}),
    '10': ModelSpec({'num_classes': 10}),
}

plane_cifar100_book = {
    '2': ModelSpec({'num_classes': 100}),
    '10': ModelSpec({'num_classes': 100}),
}

# Make these available in the global scope for the function
import sys
current_module = sys.modules[__name__]
setattr(current_module, 'resnet_book', resnet_book)
setattr(current_module, 'plane_cifar10_book', plane_cifar10_book)
setattr(current_module, 'plane_cifar100_book', plane_cifar100_book)
setattr(current_module, 'ConvNetMaker', ConvNetMaker)

ARGS = {
    'name': 'resnet110',
    'dataset': 'cifar100',
    'use_cuda': False,
}
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 SimpleResNet — a SimpleResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleResNet 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
is_resnet Ran
ran, violated the contract we drafted
imirzadeh/Teacher-Assistant-Knowledge-Distillation/model_factory.py
code served (permissive licence) · get_code("d69bff44ba26e12d")
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 bool, ndim 0, dtype family float, shape () 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 = {
    "name": "ResNet50"
}
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 bool, ndim 0, dtype family float, shape () — shape () 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 1 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
is_resnet 1 other paper
1 later
2407.05316 (2024-07)
Leveraging Topological Guidance for Improved Knowledge Distillation

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

Despite the fact that deep neural networks are powerful models and achieve appealing results on many tasks, they are too large to be deployed on edge devices like smartphones or embedded sensor nodes. There have been efforts to compress these networks, and a popular method is knowledge distillation, where a large (teacher) pre-trained network is used to train a smaller (student) network. However, in this paper, we show that the student network performance degrades when the gap between student and teacher is large. Given a fixed student network, one cannot employ an arbitrarily large teacher, or in other words, a teacher can effectively transfer its knowledge to students up to a certain size, not smaller. To alleviate this shortcoming, we introduce multi-step knowledge distillation, which employs an intermediate-sized network (teacher assistant) to bridge the gap between the student and the teacher. Moreover, we study the effect of teacher assistant size and extend the framework to multi-step distillation. Theoretical analysis and extensive experiments on CIFAR-10,100 and ImageNet datasets and on CNN and ResNet architectures substantiate the effectiveness of our proposed approach.

For agents

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

get_harvested_code_for_paper("1902.03393")
get_code_for_paper("1902.03393")
have("1902.03393")

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

curl https://syntology.ai/api/ran/1902.03393.json

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

[![Syntology run record](https://syntology.ai/api/ran/1902.03393.svg)](https://syntology.ai/paper/1902.03393)

Connect an agent — have() is free.