Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-15T22:47:49.169054Z
Paper Citation Record · LEDGER
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA
As of 19 August 2026, this Paper Citation Record lists 30 of 30 outbound references and 0 inbound Pith citation observations for arXiv:2505.06356.
A citation records a reference. It does not transfer a finding from one paper to another.
Coverage vector
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links
A source-named dated measurement, never combined with another source.
Source: cited_works
Reference resolution
30 of 30 outbound references displayed
- verified exact0
- verified fuzzy20
- unresolved10
- parse uncertain0
- malformed identifier0
- metadata mismatch0
External citation measurements
No source-named external measurement is stored.
Outbound references
Observation 663b3fed-252e-4bcd-9f2b-3b31b2a02610 · outbound
This paper cites Flamingo: a Visual Language Model for Few-Shot Learning.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Flamingo: a Visual Language Model for Few-Shot Learning
Reference 1
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.579507Z
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 cbaf101d-231b-4de9-9e3a-656974b9fcbb · outbound
This paper cites Y our vision-language model itself is a strong filter: Toward s high-quality instruction tuning with data selection, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Y our vision-language model itself is a strong filter: Toward s high-quality instruction tuning with data selection, 2024
Reference 2
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.566146Z
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 c8739de5-0843-4a1d-9212-26007803a608 · outbound
This paper cites Comm: A coherent inter- leaved image-text dataset for multimodal understanding an d generation, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Comm: A coherent inter- leaved image-text dataset for multimodal understanding an d generation, 2024
Reference 3
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.553590Z
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 571a04cb-9f43-4929-ab0f-3c26ca496045 · outbound
This paper cites PaLI: A Jointly-Scaled Multilingual Language-Image Model.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA PaLI: A Jointly-Scaled Multilingual Language-Image Model
Reference 4
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.063146Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation beb628dc-a79c-441d-b713-aeae39b9511f · outbound
This paper cites PaLI-X: On Scaling up a Multilingual Vision and Language Model.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA PaLI-X: On Scaling up a Multilingual Vision and Language Model
Reference 5
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.067724Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4b1fa9a7-9aa9-4813-9bd3-f003bb63ac61 · outbound
This paper cites Command R.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Command R
Reference 6
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.540544Z
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 4b29b4f1-c834-4464-a24f-ca449b161f2c · outbound
This paper cites Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models
Reference 7
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.076972Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 11a0ef41-deff-45d3-befd-cdea1cce37d2 · outbound
This paper cites Detoxify.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Detoxify
Reference 8
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.527155Z
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 b1b27d42-1677-40a9-80d5-d26b7f492402 · outbound
This paper cites LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models
Reference 9
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.084993Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 28400e92-8541-4288-8602-28fc2a78a483 · outbound
This paper cites Jailbreakzoo: Survey, landscapes, and horizons in jailbreaking large language an d vision-language models, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Jailbreakzoo: Survey, landscapes, and horizons in jailbreaking large language an d vision-language models, 2024
Reference 10
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.513777Z
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 72617fd2-c0ba-45ae-8ca4-8d833f511516 · outbound
This paper cites Vhelm: A holistic evaluation of vision language mod- els, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Vhelm: A holistic evaluation of vision language mod- els, 2024
Reference 11
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.500892Z
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 0016f5da-6da0-4bd6-832d-0f41ca6cdf4f · outbound
This paper cites Elite: Enhanced language-image toxicity evaluation for safety, 2025.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Elite: Enhanced language-image toxicity evaluation for safety, 2025
Reference 12
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.487453Z
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 7c52c311-f8aa-44e2-982b-3c03ae64f2f2 · outbound
This paper cites Improved Baselines with Visual Instruction Tuning, 2023.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Improved Baselines with Visual Instruction Tuning, 2023
Reference 13
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.472673Z
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 0f0d46d8-289f-45b3-80c1-1269dd77d017 · outbound
This paper cites Visual Instruction Tuning, 2023.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Visual Instruction Tuning, 2023
Reference 14
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.458294Z
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 13475d53-5d5d-486c-b399-112d1f2cb2f0 · outbound
This paper cites Mm-safetybench: A benchmark for safety eval- uation of multimodal large language models, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Mm-safetybench: A benchmark for safety eval- uation of multimodal large language models, 2024
Reference 15
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.445746Z
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 2a99a062-e29d-44f4-b738-b0a4e56e1c93 · outbound
This paper cites Towards interpreting visual infor - mation processing in vision-language models, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Towards interpreting visual infor - mation processing in vision-language models, 2024
Reference 16
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.431701Z
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 92697fc4-b12a-4c82-99dd-ea89b95b9e8a · outbound
This paper cites Kosmos-G: Generating Images in Context with Multimodal Large Language Models.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Kosmos-G: Generating Images in Context with Multimodal Large Language Models
Reference 17
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.116160Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 1647ad66-906c-4a99-940b-0121ea53c54e · outbound
This paper cites Kosmos-2: Grounding Multimodal Large Language Models to the World.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Kosmos-2: Grounding Multimodal Large Language Models to the World
Reference 18
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.120423Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 088b5361-2522-4ad9-b26a-8b90d41c74a5 · outbound
This paper cites Learn- ing Transferable Visual Models From Natural Language Su- pervision.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Learn- ing Transferable Visual Models From Natural Language Su- pervision
Reference 19
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.418176Z
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 6543f544-6704-401b-9c6b-d4d5116611c4 · outbound
This paper cites Training-free mitigation of language reasoning degradation after multimodal instruction tuning, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Training-free mitigation of language reasoning degradation after multimodal instruction tuning, 2024
Reference 20
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.404586Z
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 852ddd3e-c245-4c8d-80de-3d3a7fbc7846 · outbound
This paper cites Laion-5b: An open large-scale dataset for training next generation image-text models, 2022.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Laion-5b: An open large-scale dataset for training next generation image-text models, 2022
Reference 21
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.391510Z
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 66fd9b2b-a4bd-4769-a8c1-b8851402f569 · outbound
This paper cites From pixels to prose: A large dataset of dense image cap- tions, 2024.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA From pixels to prose: A large dataset of dense image cap- tions, 2024
Reference 22
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.378068Z
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 0e45e7a4-cd81-4d42-b2c9-df0cac1f1b5f · outbound
This paper cites RoFormer: Enhanced Transformer with Rotary Position Embedding, 2021.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA RoFormer: Enhanced Transformer with Rotary Position Embedding, 2021
Reference 23
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.141758Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e120afa9-4647-45ad-afdd-aae5e0c7bdef · outbound
This paper cites Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
Reference 24
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.145359Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 1aeacf78-2ebc-4b92-840c-ed201d70f13a · outbound
This paper cites Florence-2: Advancing a unified representation for a variet y of vision tasks.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Florence-2: Advancing a unified representation for a variet y of vision tasks
Reference 25
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.148914Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 3da6cc34-7eb3-48ab-9aca-e841e01b8453 · outbound
This paper cites Pangea: A Fully Open Multilingual Multimodal LLM for 39 Languages.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Pangea: A Fully Open Multilingual Multimodal LLM for 39 Languages
Reference 26
- Resolution
- unresolved
- no resolver link, observed 2026-08-15T22:47:49.152993Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4f8a7d9c-5f57-4835-9f93-a60c693cfe9e · outbound
This paper cites Sigmoid Loss for Language Image Pre- Training.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Sigmoid Loss for Language Image Pre- Training
Reference 27
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.348897Z
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 23d66c40-6f80-4e29-a430-85d07664c789 · outbound
This paper cites Spa-vl: A comprehensive safety preference alignment dataset for vi - sion language model, 2025.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Spa-vl: A comprehensive safety preference alignment dataset for vi - sion language model, 2025
Reference 28
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.336125Z
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 aeeabb02-1824-46e6-aa80-75e3ffa4eb70 · outbound
This paper cites Zero-shot defense against toxic images via inherent multimodal alignment in lvlms, 2025.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Zero-shot defense against toxic images via inherent multimodal alignment in lvlms, 2025
Reference 29
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.322869Z
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 53243062-5057-4e78-81f9-1e3cb69f06cd · outbound
This paper cites Un- derstanding and rectifying safety perception distortion i n vlms, 2025.
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA Un- derstanding and rectifying safety perception distortion i n vlms, 2025
Reference 30
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-08-15T22:47:49.309450Z
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.
Pith citing papers
No inbound Pith citation observations are available.