SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.12964 · 2024

Enhancing Vision-Language Few-Shot Adaptation with Negative Learning

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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; 3 failures of our instrument; 3 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/2403.12964.json · badge: /api/ran/2403.12964.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
zhangce01/dualadapter canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
FunctionStatusWhere it lives
basic_clean Ran
ran; our contract draft was wrong, not the code
zhangce01/dualadapter/clip/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("98f385d847636a3e")
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 ltpwy/MSCI, 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_pairs Ran
ran; our contract draft was wrong, not the code
zhangce01/dualadapter/clip/simple_tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
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 set. 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

torch.manual_seed(0)

# The function get_pairs takes a word parameter
# A word is represented as a tuple of symbols (variable-length strings)
# The function iterates through the word and creates pairs of consecutive symbols

word = ("hello", "world", "test")

ARGS = {
    "word": word
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
load_few_shot_feature Ran
ran, honoured the contract we drafted
zhangce01/dualadapter/utils.py
code served (permissive licence) · get_code("98959a36f76d691e")
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 tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 tempfile
import os

torch.manual_seed(0)

# Create a temporary directory to serve as cache_dir
temp_dir = tempfile.mkdtemp()

# Create mock cache files with small tensors
shots = 3
cache_keys = torch.randn(2, 4)  # batch_size=2, dim=4
cache_values = torch.randn(2, 4)  # batch_size=2, dim=4

# Save the tensors to the temporary directory
torch.save(cache_keys, os.path.join(temp_dir, f'keys_{shots}shots.pt'))
torch.save(cache_values, os.path.join(temp_dir, f'values_{shots}shots.pt'))

# Create config dict with required parameters
cfg = {
    'cache_dir': temp_dir,
    'shots': shots
}

ARGS = {
    'cfg': cfg
}
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 tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
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
load_text_feature Ran
ran, honoured the contract we drafted
zhangce01/dualadapter/utils.py
code served (permissive licence) · get_code("a44e1ca49828cc93")
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 tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 tempfile
import os

torch.manual_seed(0)

# Create a temporary directory to act as cache_dir
temp_dir = tempfile.mkdtemp()

# Create small dummy tensor files that the function will load
# These represent CLIP text feature weights
dummy_template = torch.randn(2, 4)
dummy_cupl = torch.randn(2, 4)
dummy_negative = torch.randn(2, 4)

# Save the tensors to the expected paths
torch.save(dummy_template, os.path.join(temp_dir, 'text_weights_template.pt'))
torch.save(dummy_cupl, os.path.join(temp_dir, 'text_weights_cupl.pt'))
torch.save(dummy_negative, os.path.join(temp_dir, 'text_weights_negative_template.pt'))

# Create the config dictionary
cfg = {
    'cache_dir': temp_dir
}

ARGS = {
    'cfg': cfg
}
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 tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
zhangce01/dualadapter/clip/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("9542161e9640b858")
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 ltpwy/MSCI, 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.

build_model Not yet run
unverified (harvested, no recorded run)
zhangce01/dualadapter/clip/model.py
code served (permissive licence) · get_code("6a9aff2f8e7115de")
How it ran

no invocation recorded.

cls_acc Not yet run
unverified (harvested, no recorded run)
zhangce01/dualadapter/utils.py
code served (permissive licence) · get_code("f2a0a5061e7c204e")
How it ran

no invocation recorded.

load Not yet run
unverified (harvested, no recorded run)
zhangce01/dualadapter/clip/clip.py
pointer only (licence: MIT); the stored copy get_code serves, from gaopengcuhk/Tip-Adapter, is pointer only (licence: NONE) · get_code("db64af1ba3479e8d")
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 gaopengcuhk/Tip-Adapter, 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.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 5 the same month, 6 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
get_pairs 258 other papers
116 dated earlier, 14 the same month, 120 later, 7 same year, order not determinable, 1 undated · 4 only as an assembly brick
1802.06474 (2018-02)
A Closed-form Solution to Photorealistic Image Stylization
1803.01128 (2018-03)
Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examples
1809.00125 (2018-09)
Simple Fusion: Return of the Language Model
and 255 more not named here
whitespace_clean 231 other papers
98 dated earlier, 13 the same month, 112 later, 7 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 228 more not named here
basic_clean 218 other papers
98 dated earlier, 13 the same month, 99 later, 7 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 215 more not named here
build_model 119 other papers
58 dated earlier, 6 the same month, 48 later, 5 same year, order not determinable, 2 undated
1809.11096 (2018-09)
Large Scale GAN Training for High Fidelity Natural Image Synthesis
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2104.14294 (2021-04)
Emerging Properties in Self-Supervised Vision Transformers
and 116 more not named here
load 17 other papers
6 dated earlier, 10 later, 1 same year, order not determinable
2211.16582 (2022-11)
SinDDM: A Single Image Denoising Diffusion Model
2302.00070 (2023-02)
Debiasing Vision-Language Models via Biased Prompts
and 14 more not named here
cls_acc 12 other papers
2 dated earlier, 1 the same month, 9 later · 2 only as an assembly brick
2210.01253 (2022-10)
PLOT: PROMPT LEARNING WITH OPTIMAL TRANS-PORT FOR VISION-LANGUAGE MODELS
2402.04087 (2024-02)
A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation
2403.17589 (2024-03)
Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models
and 9 more not named here

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

Large-scale pre-trained Vision-Language Models (VLMs) have exhibited impressive zero-shot performance and transferability, allowing them to adapt to downstream tasks in a data-efficient manner. However, when only a few labeled samples are available, adapting VLMs to distinguish subtle differences between similar classes in specific downstream tasks remains challenging. In this work, we propose a Simple yet effective Negative Learning approach, SimNL, to more efficiently exploit the task-specific knowledge from few-shot labeled samples. Unlike previous methods that focus on identifying a set of representative positive features defining "what is a {CLASS}", SimNL discovers a complementary set of negative features that define "what is not a {CLASS}", providing additional insights that supplement the positive features to enhance task-specific recognition capability. Further, we identify that current adaptation approaches are particularly vulnerable to potential noise in the few-shot sample set. To mitigate this issue, we introduce a plug-and-play few-shot instance reweighting technique to suppress noisy outliers and amplify clean samples for more stable adaptation. Our extensive experimental results across 15 datasets validate that the proposed SimNL outperforms existing state-of-the-art methods on both few-shot learning and domain generalization tasks while achieving competitive computational efficiency. Code is available at https://github.com/zhangce01/SimNL.

For agents

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

get_harvested_code_for_paper("2403.12964")
get_code_for_paper("2403.12964")
have("2403.12964")

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

curl https://syntology.ai/api/ran/2403.12964.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.12964.svg)](https://syntology.ai/paper/2403.12964)

Connect an agent — have() is free.