Paper Citation Record · LEDGER

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models

As of 18 August 2026, this Paper Citation Record lists 18 of 18 outbound references and 1 inbound Pith citation observation for arXiv:2505.24227.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2505.24227 v1

Coverage vector

measured 18 of 18 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T12:35:37.174665Z

measured 19 of 19 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-10T15:58:28.202606Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-05-11T09:31:05.046964Z

Reference resolution

18 of 18 outbound references displayed

verified exact0
verified fuzzy5
unresolved12
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 96877b59-a946-49e9-81a3-54fa8c3eed8d · outbound

This paper cites Controlled Caption Generation for Images Through Adversarial Attacks.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Controlled Caption Generation for Images Through Adversarial Attacks

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:34.858781Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 471cbf59-d131-4b80-8c97-d3783b4df21e · outbound

This paper cites AICAttack: Adversarial Image Captioning Attack with Attention-Based Optimization.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models AICAttack: Adversarial Image Captioning Attack with Attention-Based Optimization

Reference 8

Resolution: metadata mismatch; local_arxiv, observed 2026-08-07T12:35:37.349893Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 438c66c2-4e98-4e3b-b30a-6a4730ea0a76 · outbound

This paper cites LXMERT: Learning Cross-Modality Encoder Representations from Transformers.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models LXMERT: Learning Cross-Modality Encoder Representations from Transformers

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:36.557475Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 70145193-9325-469e-be29-ff284fd76d58 · outbound

This paper cites an unresolved cited work.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Unresolved cited work

Reference 13

Resolution: unresolved; raw_fallback, observed 2026-08-07T12:35:38.631733Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f85a36c9-787c-4b5f-8d7e-01acd4b2bcee · outbound

This paper cites During the optimization procedure, let Ii denote the generated adversarial image at the ith step.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models During the optimization procedure, let Ii denote the generated adversarial image at the ith step

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:35:38.399856Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c4cac3ac-6a5b-485c-bfd3-b6b516853df9 · outbound

This paper cites an unresolved cited work.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-07T12:35:38.025899Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation d132cbeb-ebc9-4e8a-a7a6-70314acf16b1 · outbound

This paper cites an unresolved cited work.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Unresolved cited work

Reference 17

Resolution: unresolved; raw_fallback, observed 2026-08-07T12:35:37.815825Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8413f760-3128-46a4-bb36-7b2cd7a9db1c · outbound

This paper cites APA measures the percentage of successful prediction answers among all the images.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models APA measures the percentage of successful prediction answers among all the images

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:35:37.631326Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5f2c3ffc-f350-41ea-9935-df696ddb5742 · outbound

This paper cites TASA: Deceiving Question Answering Models by Twin Answer Sentences Attack.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models TASA: Deceiving Question Answering Models by Twin Answer Sentences Attack

Reference 2005

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:34.960220Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22bc1e94-042a-402c-aa38-1f117cddb53e · outbound

This paper cites ClipCap: CLIP Prefix for Image Captioning.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models ClipCap: CLIP Prefix for Image Captioning

Reference 2012

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:36.351321Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ffba26f6-b460-4727-8216-1ab2f21a672f · outbound

This paper cites In this study, we randomly choose 1,000 images from the test set of the above datasets as clean images to craft adversarial examples.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models In this study, we randomly choose 1,000 images from the test set of the above datasets as clean images to craft adversarial examples

Reference 2014

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:35:38.234101Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f60c9835-8292-467e-adf1-c4af10ca95ea · outbound

This paper cites Transfer Attack for Bad and Good: Explain and Boost Adversarial Transferability across Multimodal Large Language Models.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Transfer Attack for Bad and Good: Explain and Boost Adversarial Transferability across Multimodal Large Language Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:35.084001Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd228681-debf-4fbd-884c-44a6458b6f6d · outbound

This paper cites A Survey on Transferability of Adversarial Examples across Deep Neural Networks.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models A Survey on Transferability of Adversarial Examples across Deep Neural Networks

Reference 2018

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:35.411237Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5699b861-47e4-4f3f-97d4-bcb9eaa6e1f2 · outbound

This paper cites S., Oh, C., and Cavallaro, A.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models S., Oh, C., and Cavallaro, A

Reference 2021

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:35:38.788221Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 87da1941-05b3-4541-bca0-b79fd8f91043 · outbound

This paper cites OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:35.568500Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94ac1984-3a7a-4856-a826-f280259693f9 · outbound

This paper cites SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:35.832321Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3c11b9b-a659-4a2e-9bf5-d6e7fd6315e6 · outbound

This paper cites Adversarial VQA: A new benchmark for evaluating the robustness of VQA models.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Adversarial VQA: A new benchmark for evaluating the robustness of VQA models

Reference 2024

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:35:38.976559Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6218c211-9f0b-4491-a79a-0356d7360607 · outbound

This paper cites Motivating the Rules of the Game for Adversarial Example Research.

Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models Motivating the Rules of the Game for Adversarial Example Research

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-07T12:35:35.259685Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation ab4d6459-03a9-46c2-b864-be24593e8fb0 · inbound

Challenging Vision-Language Models with Physically Deployable Multimodal Semantic Lighting Attacks cites this paper.

Challenging Vision-Language Models with Physically Deployable Multimodal Semantic Lighting Attacks Light as Deception: GPT-driven Natural Relighting Against Vision-Language Pre-training Models

Reference 43

Resolution: verified exact; arxiv_id, observed 2026-05-11T09:31:05.052907Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.