Papers › Decomposing and Interpreting Image Representations via Text in ViTs Beyond CLIP

Decomposing and Interpreting Image Representations via Text in ViTs Beyond CLIP

3 Jun 2024arXiv:2406.01583archive 2025-07-28

Sriram Balasubramanian, Samyadeep Basu, Soheil Feizi

Recent work has explored how individual components of the CLIP-ViT model contribute to the final representation by leveraging the shared image-text representation space of CLIP. These components, such as attention heads and MLPs, have been shown to capture distinct image features like shape, color or texture. However, understanding the role of these components in arbitrary vision transformers (ViTs) is challenging. To this end, we introduce a general framework which can identify the roles of various components in ViTs beyond CLIP. Specifically, we (a) automate the decomposition of the final representation into contributions from different model components, and (b) linearly map these contributions to CLIP space to interpret them via text. Additionally, we introduce a novel scoring function to rank components by their importance with respect to specific features. Applying our framework to various ViT variants (e.g. DeiT, DINO, DINOv2, Swin, MaxViT), we gain insights into the roles of different components concerning particular image features. These insights facilitate applications such as image retrieval using text descriptions or reference images, visualizing token importance heatmaps, and mitigating spurious correlations. We release our code to reproduce the experiments at https://github.com/SriramB-98/vit-decompose

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="2406.01583")

Code

Syntology Ran 10 of 13 code samples harvested from 1 repository linked to this paper; 3 have no recorded run. Of those that ran: 1 ran · honoured contract; 9 ran with no contract checked.

By repository: official repository: 13 samples from 1 repository, 10 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

sriramb-98/vit-decompose officialmentioned in paperpytorch 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

13 samples harvested; 10 ran; 1 honoured the contract we drafted; 3 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 · honoured contract
9ran
3unverified

Licence: 13 of the 13 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 SriramB-98/vit-decompose. “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.

collect_layer_components SriramB-98/vit-decompose/helpers/model_utils.py official repository ran no licence file found · pointer only · ad1f8ca3394f8a3e · report
concat_dict SriramB-98/vit-decompose/helpers/decompose_utils.py official repository ran no licence file found · pointer only · af8eeaf806f4ef88 · report
detach_input SriramB-98/vit-decompose/helpers/model_utils.py official repository ran no licence file found · pointer only · bb93aacaffbcc8d1 · report
dict_to_tensor SriramB-98/vit-decompose/helpers/decompose_utils.py official repository ran no licence file found · pointer only · fac353c620fc2b66 · report
extract_shape_metadata SriramB-98/vit-decompose/helpers/linear_decompose.py official repository ran no licence file found · pointer only · 1984b228b8b7a70e · report
get_exec_order sriramb-98/vit-decompose/component_ablation.py official repository ran · honoured contract no licence file found · pointer only · dd461c55cc3b4e57 · report
head_outputs SriramB-98/vit-decompose/helpers/decompose_utils.py official repository ran no licence file found · pointer only · 696c02b1ccc49929 · report
list_to_tensor SriramB-98/vit-decompose/helpers/model_utils.py official repository ran no licence file found · pointer only · daf19f974b1c8d50 · report
prod SriramB-98/vit-decompose/helpers/linear_decompose.py official repository ran fingerprinted no licence file found · pointer only · f008b9c99180b5f5 · report
split SriramB-98/vit-decompose/helpers/linear_decompose.py official repository ran no licence file found · pointer only · 5f3750bcdd0d49e0 · report
checkpoint_filter_fn SriramB-98/vit-decompose/timm/models/tiny_vit.py official repository unverified no licence file found · pointer only · 3f5d672bc6ebb458 · report
get_imgs_across_heads sriramb-98/vit-decompose/image_retrieval_from_image.py official repository unverified no licence file found · pointer only · 310ebed22b298ee3 · report
get_imgs_across_heads sriramb-98/vit-decompose/image_retrieval_from_text.py official repository unverified no licence file found · pointer only · 6236e4d25c0310db · report

Tasks

Image Retrieval

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionAttention DropoutCLIPDINODeiTDense ConnectionsDropoutFeedforward NetworkLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

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