Papers › Language Models as Black-Box Optimizers for Vision-Language Models

Language Models as Black-Box Optimizers for Vision-Language Models

12 Sep 2023CVPR 2024 1arXiv:2309.05950archive 2025-07-28

Shihong Liu, Zhiqiu Lin, Samuel Yu, Ryan Lee, Tiffany Ling, Deepak Pathak, Deva Ramanan

Vision-language models (VLMs) pre-trained on web-scale datasets have demonstrated remarkable capabilities on downstream tasks when fine-tuned with minimal data. However, many VLMs rely on proprietary data and are not open-source, which restricts the use of white-box approaches for fine-tuning. As such, we aim to develop a black-box approach to optimize VLMs through natural language prompts, thereby avoiding the need to access model parameters, feature embeddings, or even output logits. We propose employing chat-based LLMs to search for the best text prompt for VLMs. Specifically, we adopt an automatic hill-climbing procedure that converges to an effective prompt by evaluating the performance of current prompts and asking LLMs to refine them based on textual feedback, all within a conversational process without human-in-the-loop. In a challenging 1-shot image classification setup, our simple approach surpasses the white-box continuous prompting method (CoOp) by an average of 1.5% across 11 datasets including ImageNet. Our approach also outperforms both human-engineered and LLM-generated prompts. We highlight the advantage of conversational feedback that incorporates both positive and negative prompts, suggesting that LLMs can utilize the implicit gradient direction in textual feedback for a more efficient search. In addition, we find that the text prompts generated through our strategy are not only more interpretable but also transfer well across different VLM architectures in a black-box manner. Lastly, we apply our framework to optimize the state-of-the-art black-box VLM (DALL-E 3) for text-to-image generation, prompt inversion, and personalization.

PaperPDFConference PDFCodeCode 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 11 code samples harvested from 1 repository linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 11 samples from 1 repository, 8 ran (6 with no instrument failure; 2 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 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

shihongl1998/llm-as-a-blackbox-optimizer officialmentioned in paperpytorchSyntology: 11 samples, 8 ran (2 instrument-failed) · 3 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

11 samples harvested: 8 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 3 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.

2ran · our draft was wrong
6ran
3unverified

Licence: 11 of the 11 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from shihongl1998/llm-as-a-blackbox-optimizer. “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.

shihongl1998/llm-as-a-blackbox-optimizer official repository 11 samples, 8 ran (2 instrument-failed) · 3 unverified
encode_image in 131 papers auto_prompt_image_optimization.py ran · our draft was wrong no licence file found · pointer only · f41cb1a19b154297 · report fixture: signature image_path:str/0/str · returned str · level recorded 2026-09-22
get_experiment_name utils.py ran no licence file found · pointer only · 60d87e4e28792d30 · report fixture: signature args:Args/0/? · returned dict · level recorded 2026-09-24
get_message in 8 papers prompt_pool.py ran no licence file found · pointer only · 5bc725a3a530c2c4 · report fixture: signature prompt_method:str/0/str, good_templates_str:str/0/str, bad_templates_str:str/0/str, random_templates_str:str/0/str, num_templates_from_gpt:int/0/int, dataset:list/1/str · returned str · level recorded 2026-09-24
get_split_transformer engine/clip/partial_model.py ran no licence file found · pointer only · e07e33e12b2f42d7 · report fixture: signature clip_model:callable/0/?, layer_idx:int/0/int · returned Encoder · level recorded 2026-09-24
get_split_vit engine/clip/partial_model.py ran no licence file found · pointer only · 219ed3eef816ba50 · report fixture: signature model:callable/0/?, layer_idx:int/0/int · returned Encoder · level recorded 2026-09-24
get_text_encoder engine/clip/partial_model.py ran no licence file found · pointer only · 0aa8eb7ae30a6420 · report fixture: signature text_layer_idx:int/0/int, clip_model:callable/0/? · returned Encoder · level recorded 2026-09-24
load_templates_from_txt load_laioncoco_templates.py ran no licence file found · pointer only · e62bfda6e308ebbc · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-24
read_classnames engine/datasets/imagenet.py ran · our draft was wrong no licence file found · pointer only · 92f525cf6230e44e · report fixture: signature text_file:str/0/str · returned OrderedDict · level recorded 2026-09-25
ask_chatgpt_for_prompt auto_prompt_prompt_inversion.py unverified no licence file found · pointer only · 31025d8eb991afe7 · report
build_model in 469 papers engine/clip/model.py unverified no licence file found · pointer only · c47aa9e5b049a11d · report
get_init_templates utils.py unverified no licence file found · pointer only · 5ca56bc26635e339 · report

Tasks

Few-Shot Image ClassificationImage ClassificationImage GenerationText to Image GenerationText-to-Image Generationimage-classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

CLIPCoOp

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