Paper Citation Record · LEDGER

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling

As of 10 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 0 inbound Pith citation observations for arXiv:2502.00965.

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

pith.paper-citation-record.v1
2502.00965 v2

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-09T17:09:26.916926Z

measured 24 of 24 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

24 of 24 outbound references displayed

verified exact0
verified fuzzy5
unresolved17
parse uncertain0
malformed identifier0
metadata mismatch2

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 9bffacab-8da9-4c87-bcfe-362947f1ec7d · outbound

This paper cites Data Filtering Networks.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Data Filtering Networks

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.814720Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c977b14-1ae5-4930-b694-88480c28e905 · outbound

This paper cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.819582Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da12ef41-fa2a-409e-a021-33b89bf382af · outbound

This paper cites Vision-Language Pre-training: Basics, Recent Advances, and Future Trends.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Vision-Language Pre-training: Basics, Recent Advances, and Future Trends

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.824870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a60b733-af98-4b83-aada-8d63d804d106 · outbound

This paper cites Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.829582Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be150f40-1a03-43f1-873f-2eec9a0514fd · outbound

This paper cites Visual Instruction Tuning.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Visual Instruction Tuning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.838908Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 52e583d0-4c1a-4a57-a591-15db071bef13 · outbound

This paper cites Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.844565Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7173aa85-6603-4035-a46b-915379cdb175 · outbound

This paper cites Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.854146Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03148481-167a-4e91-b3ed-651a88b27d76 · outbound

This paper cites Zero-Shot Text-to-Image Generation.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Zero-Shot Text-to-Image Generation

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.859300Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc713ba7-fd57-4c3d-8156-fb3db9a093c0 · outbound

This paper cites DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.864663Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82585d86-4371-4a62-9803-c3bf8b167c9e · outbound

This paper cites OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.890649Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1924bb8f-5b51-4c80-bbda-4a0092bb44ed · outbound

This paper cites Turn Waste into Worth: Rectifying Top-$k$ Router of MoE.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Turn Waste into Worth: Rectifying Top-$k$ Router of MoE

Reference 20

Resolution: metadata mismatch; local_arxiv, observed 2026-08-09T17:09:26.993491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation d6728183-3897-4491-a6dd-cd8545816c68 · outbound

This paper cites CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.901121Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 16a50ade-f423-4e41-b014-881d15b63744 · outbound

This paper cites ST-MoE: Designing Stable and Transferable Sparse Expert Models.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling ST-MoE: Designing Stable and Transferable Sparse Expert Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.906465Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be1f5a7e-1c6f-4175-bcf0-af57ce103252 · outbound

This paper cites Table 3 summarizes the hyper-parameters for all experiments, including MoE-specific configurations and parameters for dense CLIP, sparse CLIP, and CLIP-UP.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Table 3 summarizes the hyper-parameters for all experiments, including MoE-specific configurations and parameters for dense CLIP, sparse CLIP, and CLIP-UP

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T17:09:27.269015Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 7b878faf-4901-46b7-a7b9-6e30deae155b · outbound

This paper cites We also explore the effect of adding MoE layers to only one modality while keeping the other modality fully dense.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling We also explore the effect of adding MoE layers to only one modality while keeping the other modality fully dense

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T17:09:27.252859Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 3684fb01-c0a0-45d1-a549-2a8c9cfd5f22 · outbound

This paper cites In 2009 IEEE conference on computer vision and pattern recognition , pages 248–255.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling In 2009 IEEE conference on computer vision and pattern recognition , pages 248–255

Reference 2009

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T17:09:27.298745Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 3d880762-b503-4d91-8970-3018eeb334de · outbound

This paper cites In Computer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, pages 740–755.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling In Computer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, pages 740–755

Reference 2014

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T17:09:27.283980Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation d8e821b2-48af-4d2c-8062-ee3f01a9c005 · outbound

This paper cites Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.875283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2a30d28-153b-4394-9b44-cd6a3fa9114d · outbound

This paper cites Do ImageNet Classifiers Generalize to ImageNet?.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Do ImageNet Classifiers Generalize to ImageNet?

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.869436Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation decf3f13-ba02-4a25-8e07-9200ba122b8a · outbound

This paper cites On Layer Normalization in the Transformer Architecture.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling On Layer Normalization in the Transformer Architecture

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.885274Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6afed1b1-27c8-4fb3-b00d-3db306e01d99 · outbound

This paper cites Learning Transferable Visual Models From Natural Language Supervision.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling Learning Transferable Visual Models From Natural Language Supervision

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.849221Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 750675db-14cd-4cb7-83bb-84120156c793 · outbound

This paper cites GLaM: Efficient Scaling of Language Models with Mixture-of-Experts.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling GLaM: Efficient Scaling of Language Models with Mixture-of-Experts

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-09T17:09:26.809525Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba1bdd70-dc95-4b9c-a37f-ac6ef1df080d · outbound

This paper cites In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recog- nition (CVPR).

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recog- nition (CVPR)

Reference 2023

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T17:09:27.314960Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 885737dc-3bb5-4e01-8bfd-118a96a130c7 · outbound

This paper cites MOFI: Learning Image Representations from Noisy Entity Annotated Images.

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling MOFI: Learning Image Representations from Noisy Entity Annotated Images

Reference 2024

Resolution: metadata mismatch; local_arxiv, observed 2026-08-09T17:09:27.046348Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Pith citing papers

No inbound Pith citation observations are available.