Papers › TULIP: Token-length Upgraded CLIP

TULIP: Token-length Upgraded CLIP

13 Oct 2024arXiv:2410.10034archive 2025-07-28

Ivona Najdenkoska, Mohammad Mahdi Derakhshani, Yuki M. Asano, Nanne van Noord, Marcel Worring, Cees G. M. Snoek

We address the challenge of representing long captions in vision-language models, such as CLIP. By design these models are limited by fixed, absolute positional encodings, restricting inputs to a maximum of 77 tokens and hindering performance on tasks requiring longer descriptions. Although recent work has attempted to overcome this limit, their proposed approaches struggle to model token relationships over longer distances and simply extend to a fixed new token length. Instead, we propose a generalizable method, named TULIP, able to upgrade the token length to any length for CLIP-like models. We do so by improving the architecture with relative position encodings, followed by a training procedure that (i) distills the original CLIP text encoder into an encoder with relative position encodings and (ii) enhances the model for aligning longer captions with images. By effectively encoding captions longer than the default 77 tokens, our model outperforms baselines on cross-modal tasks such as retrieval and text-to-image generation. The code repository is available at https://github.com/ivonajdenkoska/tulip.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 20 code samples harvested from 1 repository linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 5 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 20 samples from 1 repository, 13 ran (8 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-24 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 5 of the 13 samples that ran have no recorded date.

ivonajdenkoska/tulip officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 20 samples, 13 ran (5 instrument-failed) · 7 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

20 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 5 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
5ran · our draft was wrong
7ran
7unverified

Licence: 7 of the 20 samples are pointer only, meaning Syntology does not serve that copy's text. For 7 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from ivonajdenkoska/tulip. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

ivonajdenkoska/tulip official repository 20 samples, 13 ran (5 instrument-failed) · 7 unverified
natural_key in 10 papers eval_tulip.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · a677092c48744688 · report fixture: signature string_:str/0/str · returned list · level recorded, date not recorded
collate_fn in 500 papers eval/dci_long.py ran · our draft was wrong Apache-2.0 (permissive) · 09c78a125c9f58ae · report fixture: signature batch:list/1/tuple · returned tuple · level recorded 2026-09-24
convert_to_custom_text_state_dict in 59 papers open_clip/model.py ran · violated contract Apache-2.0 (permissive) here; the stored copy is pointer only · f19962ebb134b3d7 · report fixture: signature state_dict:dict/0/? · returned dict · level recorded, date not recorded
get_accuracy_t2i eval/flickr30.py ran fingerprinted Apache-2.0 (permissive) · 385a40328778e8cb · report fixture: signature text_feature:Tensor/2/float, image_feature:Tensor/2/float, k:int/0/int · returned float · level recorded 2026-09-24
get_cast_dtype in 67 papers open_clip/model.py ran · our draft was wrong Apache-2.0 (permissive) · dcd422d66b0581d8 · report fixture: signature precision:str/0/str · returned dtype · level recorded, date not recorded
get_clip_metrics eval/dci_long.py ran Apache-2.0 (permissive) · 0b9ff430cf687b70 · report fixture: signature image_features:Tensor/2/float, text_features:Tensor/2/float, logit_scale:Tensor/0/float · returned dict · level recorded 2026-09-24
get_image_feature eval/flickr30.py ran Apache-2.0 (permissive) · 0434c3049ad999d2 · report fixture: signature model:callable/0/?, processor:callable/0/?, data_path:str/0/str · returned Tensor · level recorded 2026-09-24
get_input_dtype in 30 papers open_clip/model.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · b476c8cfbf0f1f47 · report fixture: signature precision:str/0/str · returned dtype · level recorded, date not recorded
get_latest_checkpoint in 13 papers eval_tulip.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 93448c613c050973 · report fixture: signature path:str/0/str · returned str · level recorded, date not recorded
get_text_feature eval/flickr30.py ran Apache-2.0 (permissive) · 5be654577f88529b · report fixture: signature model:callable/0/?, distilled_model:callable/0/?, processor:MinimalProcessor/0/?, data_path:str/0/str · returned Tensor · level recorded 2026-09-24
prepare_inputs_for_generation in 58 papers open_clip/coca_model.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · fb651d0a97fd4d3f · report fixture: signature input_ids:Tensor/2/int, image_inputs:Tensor/3/float, past:NoneType/0/?, attention_mask:Tensor/2/int, position_ids:Tensor/2/int · returned dict · level recorded 2026-09-26
register_pooler in 46 papers open_clip/hf_model.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 2a377da4a76a2d44 · report fixture: signature cls:callable/0/? · returned type · level recorded 2026-09-26
run_urban1k_openclip eval/urban1k.py ran Apache-2.0 (permissive) · 287dbd8a0930623d · report fixture: signature model:callable/0/?, distilled_model:callable/0/?, processor:callable/0/?, data_path:str/0/str · returned dict · level recorded 2026-09-24
gather_features in 78 papers open_clip/loss.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · ddcbd45e940484ee · report
neighbour_exchange in 22 papers open_clip/loss.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · e332856e3c2fc814 · report
neighbour_exchange_bidir in 19 papers open_clip/loss.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 5b1fd364afcf3c05 · report
run_coco eval/coco.py unverified Apache-2.0 (permissive) · fef28a6f8d2d2a7a · report
run_dci_long_openclip eval/dci_long.py unverified Apache-2.0 (permissive) · 4a7520e4ce7edf46 · report
run_sharegpt4v eval/sharegpt4v.py unverified Apache-2.0 (permissive) · 365d03d6a9cc44e6 · report
run_sharegpt4v_openclip eval/sharegpt4v.py unverified Apache-2.0 (permissive) · 388875d0199845b5 · report

Tasks

Image GenerationText to Image GenerationText-to-Image Generation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

CLIPRelative Position Encodings

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections