TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models | Paper Citation Record

Observation ca505ee9-72fb-4e9a-a55d-ac784213d483 · outbound

This paper cites https : / / github.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models https : / / github

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.642258Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 572d7d2e-9b5f-491a-b792-6fbe0580eb55 · outbound

This paper cites https://github.com/ Seth-Park/RobustChangeCaptioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models https://github.com/ Seth-Park/RobustChangeCaptioning

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.651655Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 52bae144-4cda-422f-a59e-5d923c86ecf9 · outbound

This paper cites https : / / github.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models https : / / github

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.727100Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 1faf6095-33c5-4b2a-a557-8ea8225e67ae · outbound

This paper cites siglip needs registers for comparison, here’s dino-v2 with registers. it has five extra tokens for the model to work with: one cls token and four.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models siglip needs registers for comparison, here’s dino-v2 with registers. it has five extra tokens for the model to work with: one cls token and four

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.686292Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation d2e3c8b0-ba72-4386-86f9-ea8e4aba3a9a · outbound

This paper cites Quantifying Attention Flow in Transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Quantifying Attention Flow in Transformers

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.680273Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f5bf9054-d6c4-4664-a62f-1661062c4527 · outbound

This paper cites Debugging Tests for Model Explanations.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Debugging Tests for Model Explanations

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.687978Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6d8e999-7bb2-44ab-bf4f-1bf9d5dca8db · outbound

This paper cites Explaining image clas- sifiers by removing input features using generative models.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Explaining image clas- sifiers by removing input features using generative models

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.640981Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 144db66c-0baa-4de4-b9e4-2c6c7f9a1b98 · outbound

This paper cites METEOR: An auto- matic metric for MT evaluation with improved correlation with human judgments.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models METEOR: An auto- matic metric for MT evaluation with improved correlation with human judgments

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.480113Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 05f03c3c-4620-4291-aacd-3398a9afe754 · outbound

This paper cites Sam: The sensitivity of attribution methods to hyperparameters.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Sam: The sensitivity of attribution methods to hyperparameters

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.424733Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation ed4a48e5-0c7a-4dd5-bd92-f85d1f7880e5 · outbound

This paper cites Rewriting a deep generative model.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Rewriting a deep generative model

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.389107Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 8737a29c-99eb-40cf-8ece-427e7567a568 · outbound

This paper cites Understanding the role of individual units in a deep neural network.Proceedings of the National Academy of Sciences, 117(48):30071–30078, 2020.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Understanding the role of individual units in a deep neural network.Proceedings of the National Academy of Sciences, 117(48):30071–30078, 2020

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.346017Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a0c0dede-a94b-4ca4-8ba1-95007dddd4ab · outbound

This paper cites Better plain ViT baselines for ImageNet-1k.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Better plain ViT baselines for ImageNet-1k

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.735003Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c2d1d0a-e1a8-481c-b063-7463c3ddbd9e · outbound

This paper cites Vixen: Visual text comparison network for image difference captioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Vixen: Visual text comparison network for image difference captioning

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.312956Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6b30731e-e5d6-47e8-98e8-98dfdc6f40e9 · outbound

This paper cites Michigan man wrongfully accused with facial recognition urges Congress to act.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Michigan man wrongfully accused with facial recognition urges Congress to act

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.257021Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 49c6539d-a73e-4d1d-b093-0076c82a0dc1 · outbound

This paper cites an unresolved cited work.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-11T04:40:16.200732Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 14bab65c-f0d2-4059-9074-f6670e325244 · outbound

This paper cites Emerg- ing properties in self-supervised vision transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Emerg- ing properties in self-supervised vision transformers

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.152467Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a5067b96-9852-4732-b00f-91c98e40de08 · outbound

This paper cites Generic attention- model explainability for interpreting bi-modal and encoder- decoder transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Generic attention- model explainability for interpreting bi-modal and encoder- decoder transformers

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.109275Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 40341ed4-a55e-447a-a4aa-54dfcd54e1a5 · outbound

This paper cites Crossvit: Cross-attention multi-scale vision transformer for image classification.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Crossvit: Cross-attention multi-scale vision transformer for image classification

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.798524Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44d3f4f7-b95e-4f42-a53a-356718601b0b · outbound

This paper cites gscorecam: What objects is clip looking at? In Proceedings of the Asian Conference on Computer Vision , pages 1959– 1975, 2022.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models gscorecam: What objects is clip looking at? In Proceedings of the Asian Conference on Computer Vision , pages 1959– 1975, 2022

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:16.007733Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 1267bee9-71b5-4e73-b3b6-f43dbeeea742 · outbound

This paper cites Concept whitening for interpretable image recognition.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Concept whitening for interpretable image recognition

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.944748Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation af7352a7-ba5f-4632-a13f-533e7d74791e · outbound

This paper cites Blender - a 3D modelling and rendering package.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Blender - a 3D modelling and rendering package

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.901961Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 736222fb-6438-4532-8d55-203299ba9e7f · outbound

This paper cites Vision transformers need registers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Vision transformers need registers

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.855838Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation bf3f9988-9c2a-424f-a7df-7e787b8ec2ea · outbound

This paper cites An image is worth 16x16 words: Transformers for image recognition at scale.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models An image is worth 16x16 words: Transformers for image recognition at scale

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.779515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 7ce08f67-a395-4f4a-99c9-1fc86e29db59 · outbound

This paper cites Towards a multimodal framework for remote sensing image change retrieval and captioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Towards a multimodal framework for remote sensing image change retrieval and captioning

Reference 24

Resolution: verified exact; local_arxiv, observed 2026-08-11T04:40:12.372706Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation c5e931e8-c616-4c4d-be32-c7d6d330c745 · outbound

This paper cites Interpretable explana- tions of black boxes by meaningful perturbation.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Interpretable explana- tions of black boxes by meaningful perturbation

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.696518Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation fa8df934-cec8-4203-847e-8eb59d9ec776 · outbound

This paper cites Openagi: When llm meets domain experts.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Openagi: When llm meets domain experts

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.644489Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6c20fc50-e27a-419f-8cc9-e7e25106ee57 · outbound

This paper cites Clip4idc: Clip for image difference captioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Clip4idc: Clip for image difference captioning

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.589182Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 10445064-b551-4a21-9e49-c0eac24d3331 · outbound

This paper cites Wrongfully arrested man sues Detroit police over false facial recognition match.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Wrongfully arrested man sues Detroit police over false facial recognition match

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.531575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b2e26d52-3ea7-47a0-98de-c919a85fc0f7 · outbound

This paper cites Which tokens to use? investi- gating token reduction in vision transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Which tokens to use? investi- gating token reduction in vision transformers

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.417015Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6f1d1728-df68-4686-93a9-841a3a600b0a · outbound

This paper cites Flawed Facial Recognition Leads To Arrest and Jail for New Jersey Man.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Flawed Facial Recognition Leads To Arrest and Jail for New Jersey Man

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.371263Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 66abd07e-8b50-48e4-a3bf-457f03e1b8ab · outbound

This paper cites Change captioning: A new paradigm for multitem- poral remote sensing image analysis.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Change captioning: A new paradigm for multitem- poral remote sensing image analysis

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.325479Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 45c94ae9-fc16-402b-8211-7254586c5c2d · outbound

This paper cites OneDiff: A Generalist Model for Image Difference Captioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models OneDiff: A Generalist Model for Image Difference Captioning

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.928996Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b022a48-c6e7-4b77-9dd2-bb0a1b0408fb · outbound

This paper cites Summers, and Yingying Zhu.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Summers, and Yingying Zhu

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.277133Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 8937f13c-a5ff-45a9-8869-fd583805823e · outbound

This paper cites Learning to describe differences between pairs of similar images.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Learning to describe differences between pairs of similar images

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.242389Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 2ef8ce1a-999d-44fb-bdec-d0ef11d773ad · outbound

This paper cites Img-Diff: Contrastive Data Synthesis for Multimodal Large Language Models.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Img-Diff: Contrastive Data Synthesis for Multimodal Large Language Models

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.953813Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d64106d8-b0f3-4233-834b-4fe7c4677643 · outbound

This paper cites Now You See Me (CME): Concept-based Model Extraction.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Now You See Me (CME): Concept-based Model Extraction

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.963613Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7e1137a7-d478-4a28-8324-a5f2b80514da · outbound

This paper cites Vilt: Vision- and-language transformer without convolution or region su- pervision.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Vilt: Vision- and-language transformer without convolution or region su- pervision

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:10.995995Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation acb038ef-8d9f-4e25-aa68-f565ace05d34 · outbound

This paper cites Concept bottleneck models.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Concept bottleneck models

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.164075Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b78ace51-0a8e-4d00-9630-b6f806d8f6b6 · outbound

This paper cites The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.015862Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb05402c-8232-4edd-9bf2-f24441ffc57b · outbound

This paper cites Dynamic graph enhanced contrastive learning for chest x-ray report generation.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Dynamic graph enhanced contrastive learning for chest x-ray report generation

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.096854Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 70d59d8e-abcb-4198-b3eb-721d097a2876 · outbound

This paper cites Liang, C.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Liang, C

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:15.031847Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 257477fb-1220-4192-835b-30b774815ffb · outbound

This paper cites ROUGE: A package for automatic evaluation of summaries.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models ROUGE: A package for automatic evaluation of summaries

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.975150Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 5d62f66c-5431-4e21-9d37-08600d125242 · outbound

This paper cites Improved baselines with visual instruction tuning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Improved baselines with visual instruction tuning

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.927937Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 53a7c641-d834-4a9e-89ca-f778bfba4cf4 · outbound

This paper cites Interpretability Beyond Classification Output: Semantic Bottleneck Networks.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Interpretability Beyond Classification Output: Semantic Bottleneck Networks

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.077011Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a1cb517a-7d4d-4927-85b6-aa45e5449c18 · outbound

This paper cites SGDR: Stochastic gradi- ent descent with warm restarts.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models SGDR: Stochastic gradi- ent descent with warm restarts

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.084296Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 121a2486-b8b7-4d04-996d-0c0d557d5cd8 · outbound

This paper cites Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.841861Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a788c71f-1001-40b8-913b-e3b3c820acae · outbound

This paper cites Spot the difference: Difference visual ques- tion answering with residual alignment.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Spot the difference: Difference visual ques- tion answering with residual alignment

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.791939Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 020238cf-4847-4ad4-96e5-1f5fe5cbd9a5 · outbound

This paper cites Locating and editing factual associations in gpt.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Locating and editing factual associations in gpt

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.755785Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation dc535ae8-4efb-4fea-b512-aa420ea07542 · outbound

This paper cites Memory-based model editing at scale.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Memory-based model editing at scale

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.708439Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 0af686c2-2903-48ef-947b-f6676e1905a1 · outbound

This paper cites The ef- fectiveness of feature attribution methods and its correlation 11 with automatic evaluation scores.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The ef- fectiveness of feature attribution methods and its correlation 11 with automatic evaluation scores

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.661838Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation f602c512-5c6d-4331-bf93-54a9fb05c6a3 · outbound

This paper cites Improving change detection by incor- porating correspondence information.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Improving change detection by incor- porating correspondence information

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.607287Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 006609de-5b9d-41b5-855c-282ccaf6df6e · outbound

This paper cites How explainable are adversarially-robust CNNs?.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models How explainable are adversarially-robust CNNs?

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.141972Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04f704ac-538f-45c5-913c-a5ce1b336319 · outbound

This paper cites an unresolved cited work.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Unresolved cited work

Reference 53

Resolution: unresolved; raw_fallback, observed 2026-08-11T04:40:14.567033Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a2cc38e9-c70d-4f94-9025-937ff13cd304 · outbound

This paper cites Bleu: a method for automatic evaluation of machine translation.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Bleu: a method for automatic evaluation of machine translation

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.511660Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 69ca58c8-6fa3-436f-a35f-e3c9c5b7f636 · outbound

This paper cites Robust change captioning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Robust change captioning

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.175648Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3369db3a-15af-4f37-8cbd-017abb361e0c · outbound

This paper cites PEEB: Part-based image clas- sifiers with an explainable and editable language bottleneck.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models PEEB: Part-based image clas- sifiers with an explainable and editable language bottleneck

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.334751Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation c651e62f-0f01-40e2-aaf2-328d54eda638 · outbound

This paper cites Deepface-emd: Re-ranking using patch-wise earth mover’s distance improves out-of- distribution face identification.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Deepface-emd: Re-ranking using patch-wise earth mover’s distance improves out-of- distribution face identification

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.237969Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation fcc54815-77c0-48e6-9886-1897cb300b12 · outbound

This paper cites Fast and interpretable face identification for out-of- distribution data using vision transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Fast and interpretable face identification for out-of- distribution data using vision transformers

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.201490Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation d099e3ea-c21e-4983-a3de-738ec096d427 · outbound

This paper cites Describing and localizing multiple changes with transform- ers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Describing and localizing multiple changes with transform- ers

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.144746Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a575f53f-daeb-4c8f-a819-4e6801a15dcd · outbound

This paper cites Learning transferable visual models from natural language supervi- sion.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Learning transferable visual models from natural language supervi- sion

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.268842Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 455668d7-94eb-4261-be5f-09a846fc08ae · outbound

This paper cites The new lawsuit that shows facial recog- nition is officially a civil rights issue.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The new lawsuit that shows facial recog- nition is officially a civil rights issue

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:14.035972Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6ffd5800-7f6c-449c-96e5-b67d3897a2f4 · outbound

This paper cites The change you want to see.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The change you want to see

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.292494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c99281f6-7dd6-4ec6-b84d-d9af97f4911b · outbound

This paper cites Explaining deep neural networks and beyond: A review of methods and applications.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Explaining deep neural networks and beyond: A review of methods and applications

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.970533Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 7767697d-892a-4b99-8615-ab0876f80a16 · outbound

This paper cites The pandemic is testing the limits of face recogni- tion.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The pandemic is testing the limits of face recogni- tion

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.917424Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 4fd33787-fdc1-4522-9c73-43ca0576e1c4 · outbound

This paper cites Reclip: A strong zero-shot baseline for referring expression compre- hension.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Reclip: A strong zero-shot baseline for referring expression compre- hension

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.874300Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 25100f87-6b2a-4528-a6f4-e83acbbe0c03 · outbound

This paper cites The stvchrono dataset: Towards continuous change recognition in time.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models The stvchrono dataset: Towards continuous change recognition in time

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.804341Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 4a825cc5-9115-48bf-8852-d3b040e7d297 · outbound

This paper cites Resolution-robust large mask inpainting with fourier convolutions.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Resolution-robust large mask inpainting with fourier convolutions

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.740485Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 8c1f87f8-280c-4bdc-81c4-fe8e5ed133f8 · outbound

This paper cites Expressing visual relationships via language.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Expressing visual relationships via language

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.678333Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 3aaf7735-ca9e-435b-836a-048e6344c70e · outbound

This paper cites Eyes wide shut? exploring the visual shortcomings of multimodal llms.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Eyes wide shut? exploring the visual shortcomings of multimodal llms

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.623663Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 0f2f387e-adac-44f4-8843-d400cb5a0b6d · outbound

This paper cites Attention is all you need.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Attention is all you need

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.455191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 8335ec1d-9c70-4ca6-a125-62c81907d01a · outbound

This paper cites Cider: Consensus-based image description evalua- tion.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Cider: Consensus-based image description evalua- tion

Reference 71

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.294760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation e397a496-d379-4258-8df5-f49192b60329 · outbound

This paper cites Learning bottleneck concepts in image classification.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Learning bottleneck concepts in image classification

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.214750Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a3dd02bc-8e3e-41ef-909c-d3d422e36bf5 · outbound

This paper cites Co-attention for conditioned image matching.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Co-attention for conditioned image matching

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.124760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation d5f7cd9f-531a-4181-9fb7-a3ae3bfa65dc · outbound

This paper cites L2C: Describing visual differences needs semantic under- 12 standing of individuals.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models L2C: Describing visual differences needs semantic under- 12 standing of individuals

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:13.033000Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation e72e372e-33ec-4ddf-9756-728d21cf47d5 · outbound

This paper cites Image difference captioning with pre-training and contrastive learning.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Image difference captioning with pre-training and contrastive learning

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:12.974961Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 380f9dac-d355-4e39-a4b4-e42913e5fef6 · outbound

This paper cites Scaling vision transformers.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Scaling vision transformers

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.850919Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fd6d6b9c-1bba-4c43-aac2-e3c8e67302c6 · outbound

This paper cites Top-down neu- ral attention by excitation backprop.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Top-down neu- ral attention by excitation backprop

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:12.891960Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 7c33fa14-d549-4911-9c57-9ed1856c1fe4 · outbound

This paper cites BERTScore: Evaluating Text Generation with BERT.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models BERTScore: Evaluating Text Generation with BERT

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.898842Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6fb05bac-2db6-426f-8c98-0f79bd294d30 · outbound

This paper cites Learning deep features for discrimina- tive localization.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Learning deep features for discrimina- tive localization

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.912157Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 06b2ccbf-3274-4bc7-bc81-c6a38ecabb64 · outbound

This paper cites Implementation details A.1.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Implementation details A.1

Reference 2016

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:40:12.784863Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 5a9f7c76-530c-4161-89d7-1f81053b20cc · outbound

This paper cites an unresolved cited work.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Unresolved cited work

Reference 2019

Resolution: parse uncertain; raw_fallback, observed 2026-08-11T04:40:14.404282Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 95fc221d-0d70-49d9-b8ec-01e9cebb77ec · outbound

This paper cites an unresolved cited work.

TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models Unresolved cited work

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-11T04:40:11.214913Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers