SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.09519 · ECCV · 2022

Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification

Hongsheng Li, Wei Zhang, Jifeng Dai, Yu Qiao, Renrui Zhang, Peng Gao, Kunchang Li, Rongyao Fang

arXiv · PDF · Open in the Atlas

Code that ran

3 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 0 have no contract verdict; 3 failures 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/2207.09519.json · badge: /api/ran/2207.09519.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
ArsenalCheng/Meta-Adapter community
Papers with Code archive
1 ran · 0 no recorded run
gaopengcuhk/tip-adapter marked official in Papers with Code
Papers with Code archive
1 ran · 0 no recorded run
opengvlab/cafo community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
build_cache_model Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ArsenalCheng/Meta-Adapter/clip/utils.py
pointer only (licence: NONE) · get_code("c9e27d8ece0c4f8d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ArsenalCheng/Meta-Adapter 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_cache_model Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gaopengcuhk/tip-adapter/utils.py
pointer only (licence: NONE) · get_code("e57ad1054530e142")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gaopengcuhk/tip-adapter 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_clip_cache_model Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
opengvlab/cafo/utils.py
code served (permissive licence) · get_code("c48d589ad9da7dec")
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 tuple. 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
import tempfile
import os

torch.manual_seed(0)

# Create a minimal CLIP-like model that has encode_image method
class MockCLIPModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.projection = nn.Linear(4, 8)
    
    def encode_image(self, images):
        # images shape: (batch, 4)
        # return shape: (batch, 8)
        return self.projection(images)

# Create a minimal data loader as a list of (images, targets) tuples
def create_train_loader_cache(num_batches=2, batch_size=2):
    loader = []
    for _ in range(num_batches):
        images = torch.randn(batch_size, 4)
        targets = torch.randint(0, 3, (batch_size,))
        loader.append((images, targets))
    return loader

# Create temporary directory for cache
temp_dir = tempfile.mkdtemp()

# Create config dict
cfg = {
    'load_cache': False,
    'augment_epoch': 2,
    'shots': 5,
    'cache_dir': temp_dir
}

# Create clip model
clip_model = MockCLIPModel()

# Create train loader
train_loader_cache = create_train_loader_cache(num_batches=2, batch_size=2)

ARGS = {
    'cfg': cfg,
    'clip_model': clip_model,
    'train_loader_cache': train_loader_cache
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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

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 (0 with at least one paper dated earlier, 0 the same month, 1 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
build_cache_model
brick of this paper's assembly: entry
1 other paper
1 later
2603.24181 (2026-03)
Unlocking Few-Shot Capabilities in LVLMs via Prompt Conditioning and Head Selection

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

Contrastive Vision-Language Pre-training, known as CLIP, has provided a new paradigm for learning visual representations using large-scale image-text pairs. It shows impressive performance on downstream tasks by zero-shot knowledge transfer. To further enhance CLIP's adaption capability, existing methods proposed to fine-tune additional learnable modules, which significantly improves the few-shot performance but introduces extra training time and computational resources. In this paper, we propose a Training-free adaption method for CLIP to conduct few-shot classification, termed as Tip-Adapter, which not only inherits the training-free advantage of zero-shot CLIP but also performs comparably to those training-required approaches. Tip-Adapter constructs the adapter via a key-value cache model from the few-shot training set, and updates the prior knowledge encoded in CLIP by feature retrieval. On top of that, the performance of Tip-Adapter can be further boosted to be state-of-the-art on ImageNet by fine-tuning the cache model for 10× fewer epochs than existing methods, which is both effective and efficient. We conduct extensive experiments of few-shot classification on 11 datasets to demonstrate the superiority of our proposed methods.

For agents

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

get_harvested_code_for_paper("2207.09519")
get_code_for_paper("2207.09519")
have("2207.09519")

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

curl https://syntology.ai/api/ran/2207.09519.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.09519.svg)](https://syntology.ai/paper/2207.09519)

Connect an agent — have() is free.