SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.00700 · 2023

WeGeFT: Weight-Generative Fine-Tuning for Multi-Faceted Efficient Adaptation of Large Models

arXiv · PDF · Open in the Atlas

Code that ran

12 of 16 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 10 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/2312.00700.json · badge: /api/ran/2312.00700.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
savadikarc/gift canonical (classifier label)
classifier basis: census
12 ran · 4 no recorded run
FunctionStatusWhere it lives
SeqToSeqEncode Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("45db30ac52f3cd6d")
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 rainstonee/CDKA, 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.

cache_to_disk Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/data.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("1d0db22ae067b53c")
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 rainstonee/CDKA, 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.

causalLMEncode Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("38de4191703bb96a")
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 rainstonee/CDKA, 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.

extract_num Ran
ran, honoured the contract we drafted
savadikarc/gift/language_modeling/math_code_instruct/eval_gsm8k.py
code served (permissive licence) · get_code("9221df7ba44d3b36")
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 int, ndim 0, dtype family int, 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 re

def extract_num(text):
    pattern = '####\\s*(\\d+)'
    match = re.search(pattern, text)
    if match:
        result = match.group(1)
    else:
        print(text)
        result = ''
    try:
        return int(result.replace(',', ''))
    except:
        print(f"'{result}' can't be converted")
        return 0


ARGS = {
    'text': '#### 42'
}
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 int, ndim 0, dtype family int, 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
find_all_linear_modules Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/run_exp.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("65f3cc97d3e5034f")
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 rainstonee/CDKA, 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.

find_hidden_state_size Ran
ran, honoured the contract we drafted
savadikarc/gift/language_modeling/math_code_instruct/run_exp.py
code served (permissive licence) · get_code("4ce9d0bdc3c04522")
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 int, ndim 0, dtype family int, 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
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 int, ndim 0, dtype family int, 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
format_example Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/eval_mmlu.py
code served (permissive licence) · get_code("dec15cc75848a98e")
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 pandas as pd
import numpy as np

# Create a small DataFrame with the structure expected by format_example
# The function expects:
# - Column 0: prompt text
# - Columns 1 to k: answer choices (k = df.shape[1] - 2)
# - Column k+1: the correct answer

# Let's create a DataFrame with 3 answer choices (k=3), so 5 columns total
data = {
    0: ['What is 2+2?'],
    1: ['3'],
    2: ['4'],
    3: ['5'],
    4: ['B']
}

df = pd.DataFrame(data)

ARGS = {
    'df': df,
    'idx': 0,
    'include_answer': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
format_subject Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/eval_mmlu.py
pointer only (licence: MIT); the stored copy get_code serves, from ShangDataLab/Deep-Contam, is pointer only (licence: NONE) · get_code("793a8b6496415457")
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 ShangDataLab/Deep-Contam, 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.

gen_prompt Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/eval_mmlu.py
code served (permissive licence) · get_code("04fd5045fce0c33a")
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 pandas as pd
import numpy as np

# Create a small DataFrame that mimics the expected structure
# Columns: [question, choice_A, choice_B, choice_C, choice_D, answer]
data = {
    0: ['What is 2+2?', 'This is option A', 'This is option B', 'This is option C', 'This is option D', 'A'],
    1: ['What is the capital of France?', 'London', 'Berlin', 'Paris', 'Madrid', 'C'],
    2: ['What is 5*3?', '10', '15', '20', '25', 'B'],
}
train_df = pd.DataFrame(data).T
train_df.columns = ['question', 'choice_0', 'choice_1', 'choice_2', 'choice_3', 'answer']

ARGS = {
    'train_df': train_df,
    'subject': 'high_school_biology',
    'k': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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_forward_hook Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/logTrainer.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("fd33ad9ae547c836")
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 rainstonee/CDKA, 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.

get_module Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/lora_plus.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("ef1ba16faaacbb49")
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 rainstonee/CDKA, 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.

post_process Ran
ran; our contract draft was wrong, not the code
savadikarc/gift/language_modeling/math_code_instruct/eval_humaneval.py
pointer only (licence: MIT); the stored copy get_code serves, from rainstonee/CDKA, is pointer only (licence: NONE) · get_code("48720117c27620bc")
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 rainstonee/CDKA, 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.

create_loraplus_optimizer Not yet run
unverified (harvested, no recorded run)
savadikarc/gift/language_modeling/math_code_instruct/lora_plus.py
code served (permissive licence) · get_code("0921ef036f2219f1")
How it ran

no invocation recorded.

load_flan_v2 Not yet run
unverified (harvested, no recorded run)
savadikarc/gift/language_modeling/math_code_instruct/data.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyElements/Bi-LoRA, is pointer only (licence: NONE) · get_code("1b37d30c5eaded49")
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 CrazyElements/Bi-LoRA, 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. No invocation recorded.

load_meta_math Not yet run
unverified (harvested, no recorded run)
savadikarc/gift/language_modeling/math_code_instruct/data.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyElements/Bi-LoRA, is pointer only (licence: NONE) · get_code("963559e86ce91fd8")
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 CrazyElements/Bi-LoRA, 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. No invocation recorded.

reinit_lora Not yet run
unverified (harvested, no recorded run)
savadikarc/gift/language_modeling/math_code_instruct/run_exp.py
code served (permissive licence) · get_code("e70a2129bcf0f57c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 12 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 12 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
gen_prompt 10 other papers
3 dated earlier, 1 the same month, 6 later
2009.03300 (2020-09)
Measuring Massive Multitask Language Understanding
2305.17306 (2023-05)
Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance
2310.05736 (2023-10)
LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models
and 7 more not named here
get_module 4 other papers
4 later
2402.12354 (2024-02)
LoRA+: Efficient Low Rank Adaptation of Large Models
2407.05000 (2024-07)
LoRA-GA: Low-Rank Adaptation with Gradient Approximation
2410.02247 (2024-10)
Theoretical Insights into Fine-Tuning Attention Mechanism: Generalization and Optimization
and 1 more not named here
SeqToSeqEncode 2 other papers
2 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
cache_to_disk 2 other papers
2 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
causalLMEncode 2 other papers
2 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
find_all_linear_modules 2 other papers
2 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
post_process 2 other papers
2 later
2408.04556 (2024-08)
BA-LoRA: Bias-Alleviating Low-Rank Adaptation to Mitigate Catastrophic Inheritance in Large Language Models
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
create_loraplus_optimizer 1 other paper
1 later
2407.05000 (2024-07)
LoRA-GA: Low-Rank Adaptation with Gradient Approximation
format_subject 1 other paper
1 later
2406.13236 (2024-06)
Data Contamination Can Cross Language Barriers
get_forward_hook 1 other paper
1 later
2602.01267 (2026-02)
Diving into Kronecker Adapters: Component Design Matters
load_flan_v2 1 other paper
1 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models
load_meta_math 1 other paper
1 later
2508.19564 (2025-08)
Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models

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

Fine-tuning large pretrained Transformer models can focus on either introducing a small number of new learnable parameters (parameter efficiency) or editing representations of a small number of tokens using lightweight modules (representation efficiency). While the pioneering method LoRA (Low-Rank Adaptation) inherently balances parameter, compute, and memory efficiency, many subsequent variants trade off compute and memory efficiency and/or performance to further reduce fine-tuning parameters. To address this limitation and unify parameter-efficient and representation-efficient fine-tuning, we propose Weight-Generative Fine-Tuning (WeGeFT, pronounced wee-gift), a novel approach that learns to generate fine-tuning weights directly from the pretrained weights. WeGeFT employs a simple low-rank formulation consisting of two linear layers, either shared across multiple layers of the pretrained model or individually learned for different layers. This design achieves multi-faceted efficiency in parameters, representations, compute, and memory, while maintaining or exceeding the performance of LoRA and its variants. Extensive experiments on commonsense reasoning, arithmetic reasoning, instruction following, code generation, and visual recognition verify the effectiveness of our proposed WeGeFT. Our code is available at https://github.com/savadikarc/wegeft

For agents

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

get_harvested_code_for_paper("2312.00700")
get_code_for_paper("2312.00700")
have("2312.00700")

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

curl https://syntology.ai/api/ran/2312.00700.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.00700.svg)](https://syntology.ai/paper/2312.00700)

Connect an agent — have() is free.