Papers › PEEB: Part-based Image Classifiers with an Explainable and Editable Language Bottleneck

PEEB: Part-based Image Classifiers with an Explainable and Editable Language Bottleneck

8 Mar 2024arXiv:2403.05297archive 2025-07-28

Thang M. Pham, Peijie Chen, Tin Nguyen, Seunghyun Yoon, Trung Bui, Anh Totti Nguyen

CLIP-based classifiers rely on the prompt containing a {class name} that is known to the text encoder. Therefore, they perform poorly on new classes or the classes whose names rarely appear on the Internet (e.g., scientific names of birds). For fine-grained classification, we propose PEEB - an explainable and editable classifier to (1) express the class name into a set of text descriptors that describe the visual parts of that class; and (2) match the embeddings of the detected parts to their textual descriptors in each class to compute a logit score for classification. In a zero-shot setting where the class names are unknown, PEEB outperforms CLIP by a huge margin (~10x in top-1 accuracy). Compared to part-based classifiers, PEEB is not only the state-of-the-art (SOTA) on the supervised-learning setting (88.80% and 92.20% accuracy on CUB-200 and Dogs-120, respectively) but also the first to enable users to edit the text descriptors to form a new classifier without any re-training. Compared to concept bottleneck models, PEEB is also the SOTA in both zero-shot and supervised-learning settings.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2403.05297")

Code

Syntology Ran 13 of 21 code samples harvested from 1 repository linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · our draft was wrong; 12 ran with no contract checked.

By repository: official repository: 21 samples from 1 repository, 13 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

anguyen8/peeb officialmentioned in papermentioned on GitHubpytorchMIT 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

21 samples harvested; 13 ran; 0 honoured the contract we drafted; 8 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · our draft was wrong
12ran
8unverified

Licence: 0 of the 21 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 anguyen8/peeb. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

bound_boxes anguyen8/peeb/src/precompute_boxes.py official repository ran MIT (permissive) · a1dd2113af7b709d · report
check_exist anguyen8/peeb/src/precompute_boxes.py official repository ran MIT (permissive) · 2ff51a5d9e6dcdf3 · report
generate_prompt_sachit anguyen8/peeb/src/chatgpt.py official repository ran fingerprinted MIT (permissive) · 8fa582c5f6fddaa6 · report
get_data_split anguyen8/peeb/src/train_owlvit.py official repository ran MIT (permissive) · 0988460641132af1 · report
get_prompt_list anguyen8/peeb/src/prompt_engineering.py official repository ran MIT (permissive) · 61d607c1eb3ac510 · report
img_transform anguyen8/peeb/src/utils.py official repository ran MIT (permissive) · 924c3ad69ec6a8fa · report
load_descriptions anguyen8/peeb/src/utils.py official repository ran MIT (permissive) · 154c00edfc8d8524 · report
nested_tensor_from_tensor_list anguyen8/peeb/src/owlvit_cls.py official repository ran MIT (permissive) · d3faa0ed46f2e436 · report
nested_tensor_from_tensor_list anguyen8/peeb/src/owlvit_inference.py official repository ran MIT (permissive) · e54dd17857816dda · report
rgb_to_hex anguyen8/peeb/src/misc/peeb_plots.py official repository ran MIT (permissive) · 334d1b6d7376281e · report
sigmoid_focal_loss anguyen8/peeb/src/owlvit_cls.py official repository ran fingerprinted MIT (permissive) · 695ff74df95375df · report
wordify anguyen8/peeb/src/utils.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · 1812ba359126089a · report
xy_in_box anguyen8/peeb/src/precompute_boxes.py official repository ran MIT (permissive) · dd6b921da641c598 · report
add_extra_negatives anguyen8/peeb/src/train_owlvit.py official repository unverified MIT (permissive) · 21070f8951fe8f1f · report
dice_loss anguyen8/peeb/src/owlvit_cls.py official repository unverified MIT (permissive) · ca18559102973d9c · report
fig_to_pil anguyen8/peeb/src/misc/peeb_plots.py official repository unverified MIT (permissive) · 5088107d67670278 · report
get_exp anguyen8/peeb/src/misc/peeb_plots.py official repository unverified MIT (permissive) · 363a6138162c638c · report
get_hierarchy_class_name anguyen8/peeb/src/prompt_engineering.py official repository unverified MIT (permissive) · d5fd0ce64f775b84 · report
get_parent_nodes anguyen8/peeb/src/prompt_engineering.py official repository unverified MIT (permissive) · 364cc7215ea1dc81 · report
num_tokens_from_messages anguyen8/peeb/src/chatgpt.py official repository unverified MIT (permissive) · 6d1cc911c7acc3cc · report
test_chatgpt anguyen8/peeb/src/chatgpt.py official repository unverified MIT (permissive) · b10b022a231102b0 · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

CLIPSET

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