Paper Citation Record · LEDGER

On the Limits of Token Reduction for Efficient Unified Vision Language Training

As of 23 July 2026, this Paper Citation Record lists 47 of 47 outbound references and 0 inbound Pith citation observations for arXiv:2606.01503.

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

pith.paper-citation-record.v1
2606.01503 v1

Coverage vector

measured 47 of 47 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-28T16:57:56.333595Z

measured 47 of 47 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-07-20T06:30:07.809122+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

47 of 47 outbound references displayed

verified exact6
verified fuzzy0
unresolved41
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation bbfd7d18-a573-4d64-a687-e89dd65c2022 · outbound

This paper cites Token merging: Your vit but faster, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Token merging: Your vit but faster, 2023

Reference 1

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5182b6d0-8fae-42f7-8347-08ec0a408930 · outbound

This paper cites Sharegpt4v: Improving large multi-modal models with better captions.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Sharegpt4v: Improving large multi-modal models with better captions

Reference 2

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation faa7894c-0022-48b4-a455-4aed482fb906 · outbound

This paper cites An image is worth 1/2 tokens after layer 2: Plug-and-play inference acceleration for large vision-language models, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training An image is worth 1/2 tokens after layer 2: Plug-and-play inference acceleration for large vision-language models, 2024

Reference 3

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 770372eb-6b43-4e78-8d0e-72b8116ef617 · outbound

This paper cites InstructBLIP: Towards general-purpose vision-language models with instruction tuning.

On the Limits of Token Reduction for Efficient Unified Vision Language Training InstructBLIP: Towards general-purpose vision-language models with instruction tuning

Reference 4

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea0ef4b0-f621-4a27-99ac-9cade3a0fa49 · outbound

This paper cites The llama 3 herd of models.arXiv e-prints, pages arXiv–2407,.

On the Limits of Token Reduction for Efficient Unified Vision Language Training The llama 3 herd of models.arXiv e-prints, pages arXiv–2407,

Reference 5

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58521017-b486-4c5d-9016-065457d67803 · outbound

This paper cites Taming transformers for high-resolution image synthesis, 2021.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Taming transformers for high-resolution image synthesis, 2021

Reference 6

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44ff07d6-1c22-4cf0-9af8-e9d4234b7a7a · outbound

This paper cites Scaling rectified flow trans- formers for high-resolution image synthesis, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Scaling rectified flow trans- formers for high-resolution image synthesis, 2024

Reference 7

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 38cf4303-dc7f-441a-8193-7b90a452f987 · outbound

This paper cites MME: A comprehensive evaluation benchmark for multimodal large language models.

On the Limits of Token Reduction for Efficient Unified Vision Language Training MME: A comprehensive evaluation benchmark for multimodal large language models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4db44839-f96d-4531-8039-1693ce4b100d · outbound

This paper cites Making llama see and draw with seed tokenizer, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Making llama see and draw with seed tokenizer, 2023

Reference 9

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8bcef66-4e17-4cc4-a754-1eb5699ce92a · outbound

This paper cites When Attention Sink Emerges in Language Models: An Empirical View.

On the Limits of Token Reduction for Efficient Unified Vision Language Training When Attention Sink Emerges in Language Models: An Empirical View

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-06-28T17:02:24.305996Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation f4be08c8-aa4a-428b-8a45-1e8138c42d1a · outbound

This paper cites Denoising diffu- sion probabilistic models, 2020.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Denoising diffu- sion probabilistic models, 2020

Reference 11

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 157e9bb7-b29d-4280-af17-4e1a4c143bc3 · outbound

This paper cites Matryoshka query trans- former for large vision-language models, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Matryoshka query trans- former for large vision-language models, 2024

Reference 12

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed59ce8f-b34b-40a4-8569-5e2f1db43333 · outbound

This paper cites Hudson and Christopher D.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Hudson and Christopher D

Reference 13

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation afdfbab7-b92a-409a-9f15-ace553f9f1a4 · outbound

This paper cites Unified language-vision pretraining in llm with dynamic discrete visual tokenization, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Unified language-vision pretraining in llm with dynamic discrete visual tokenization, 2024

Reference 14

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42b01c53-b1aa-48ea-879d-453a84b3f40c · outbound

This paper cites Seed-bench: Benchmarking multimodal llms with generative comprehension, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Seed-bench: Benchmarking multimodal llms with generative comprehension, 2023

Reference 15

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 89ad5c31-298a-4fe0-8615-788fbc0fa72e · outbound

This paper cites Playground v2.5: Three insights towards enhancing aesthetic quality in text-to-image genera- tion, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Playground v2.5: Three insights towards enhancing aesthetic quality in text-to-image genera- tion, 2024

Reference 16

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98055ec9-bb71-4cd4-b80f-211c61d6739f · outbound

This paper cites Tokenpacker: Efficient visual projector for multimodal llm, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Tokenpacker: Efficient visual projector for multimodal llm, 2024

Reference 17

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff3921b5-542c-46d0-9329-ce3915b223b9 · outbound

This paper cites Evaluating object hallucination in large vision-language models, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Evaluating object hallucination in large vision-language models, 2023

Reference 18

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 57a49fa9-51e9-4bd3-beb3-457280b55024 · outbound

This paper cites Llama-vid: An image is worth 2 tokens in large language models, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Llama-vid: An image is worth 2 tokens in large language models, 2023

Reference 19

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5eef721-4e82-4b5d-8470-f6b1b6d05811 · outbound

This paper cites Boosting multimodal large language models with visual to- kens withdrawal for rapid inference, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Boosting multimodal large language models with visual to- kens withdrawal for rapid inference, 2025

Reference 20

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb79f693-dbb4-4cf6-ab19-5298c61918be · outbound

This paper cites Improved baselines with visual instruction tuning, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Improved baselines with visual instruction tuning, 2023

Reference 21

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6c014e12-a6f6-4551-b5aa-80c8205add77 · outbound

This paper cites Visual instruction tuning.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Visual instruction tuning

Reference 22

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 088db68e-ca28-4070-b332-7c9a29264fb8 · outbound

This paper cites World model on million-length video and language with blockwise ringattention.

On the Limits of Token Reduction for Efficient Unified Vision Language Training World model on million-length video and language with blockwise ringattention

Reference 23

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c979e0a2-1a8d-47ba-90bc-5afa477b54bd · outbound

This paper cites Cheap and quick: Efficient vision- language instruction tuning for large language models, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Cheap and quick: Efficient vision- language instruction tuning for large language models, 2023

Reference 24

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8922ebe-c7cf-4373-a6e2-f50dcad78f6e · outbound

This paper cites Unitok: A uni- fied tokenizer for visual generation and understanding, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Unitok: A uni- fied tokenizer for visual generation and understanding, 2025

Reference 25

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bbe35b9d-02d7-4e02-a0e8-9838de9541ca · outbound

This paper cites Janusflow: Harmonizing autoregression and rectified flow for unified multimodal understanding and generation, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Janusflow: Harmonizing autoregression and rectified flow for unified multimodal understanding and generation, 2024

Reference 26

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce0a3167-fc2c-44e5-aeae-5cdc2e55d129 · outbound

This paper cites Learning transferable visual models from natural language supervision, 2021.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Learning transferable visual models from natural language supervision, 2021

Reference 27

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 591136e9-8ab0-4bb3-aecd-f36072565446 · outbound

This paper cites Dynamicvit: Efficient vision transformers with dynamic token sparsification.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Dynamicvit: Efficient vision transformers with dynamic token sparsification

Reference 28

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 785be7f9-7aa4-4659-93e8-89dcb2eaaf4a · outbound

This paper cites High-resolution image syn- thesis with latent diffusion models, 2022.

On the Limits of Token Reduction for Efficient Unified Vision Language Training High-resolution image syn- thesis with latent diffusion models, 2022

Reference 29

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7be0623-05e1-4c85-9800-da718381a4ed · outbound

This paper cites Llava-prumerge: Adaptive token reduction for efficient large multimodal models.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Llava-prumerge: Adaptive token reduction for efficient large multimodal models

Reference 30

Resolution: verified exact; arxiv_id, observed 2026-06-28T17:02:24.309040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation 37ab29a8-aa8c-4d0a-9d2b-2eea94331399 · outbound

This paper cites Journeydb: A benchmark for generative im- age understanding.Advances in neural information process- ing systems, 36:49659–49678, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Journeydb: A benchmark for generative im- age understanding.Advances in neural information process- ing systems, 36:49659–49678, 2023

Reference 31

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fddad9bf-4462-490c-a628-42401bd01eef · outbound

This paper cites Chameleon: Mixed-modal early-fusion foundation models, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Chameleon: Mixed-modal early-fusion foundation models, 2024

Reference 32

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5698df56-b8a2-4c2f-8be6-58ea4f5cf83e · outbound

This paper cites Visual autoregressive modeling: Scalable image generation via next-scale prediction, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Visual autoregressive modeling: Scalable image generation via next-scale prediction, 2024

Reference 33

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 066be97a-8f72-4f14-bf20-9d31fc2136d2 · outbound

This paper cites Llama: Open and efficient foundation lan- guage models, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Llama: Open and efficient foundation lan- guage models, 2023

Reference 34

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98f9b6ff-325a-46dc-84d5-ef746b640c5d · outbound

This paper cites Neural discrete representation learning,.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Neural discrete representation learning,

Reference 35

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f816c86-b247-4de3-8ea8-20d3a015286c · outbound

This paper cites Emu3: Next-token prediction is all you need, 2024.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Emu3: Next-token prediction is all you need, 2024

Reference 36

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ab1316d-da58-49bc-8104-3f5b5c10c13e · outbound

This paper cites Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-06-28T17:02:24.303829Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation 510d55b9-c96c-44f8-8248-8422ac28ebf1 · outbound

This paper cites Liquid: Language Models are Scalable and Unified Multi-modal Generators.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Liquid: Language Models are Scalable and Unified Multi-modal Generators

Reference 38

Resolution: verified exact; arxiv_id, observed 2026-06-28T17:02:24.295561Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation 6761f90c-f3e8-45aa-9e7c-f7abb418ae2c · outbound

This paper cites VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation.

On the Limits of Token Reduction for Efficient Unified Vision Language Training VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation

Reference 39

Resolution: verified exact; local_arxiv, observed 2026-06-28T17:02:24.298245Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation af1e59da-4e95-4b36-a7f6-51e418c045df · outbound

This paper cites Efficient streaming language models with attention sinks.arXiv, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Efficient streaming language models with attention sinks.arXiv, 2023

Reference 40

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85ede03a-5ba7-4186-8ab4-3cffc56e2544 · outbound

This paper cites Show-o: One Single Transformer to Unify Multimodal Understanding and Generation.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Show-o: One Single Transformer to Unify Multimodal Understanding and Generation

Reference 41

Resolution: verified exact; local_arxiv, observed 2026-06-28T17:02:24.300518Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-20T06:30:07.809122+00:00.

Observation 248ee663-a3b4-4695-83a8-cd9e1cb45e9e · outbound

This paper cites Scaling autoregressive multi-modal mod- els: Pretraining and instruction tuning, 2023.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Scaling autoregressive multi-modal mod- els: Pretraining and instruction tuning, 2023

Reference 42

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5bf6ecac-c1a9-4436-afa4-14e1ea516849 · outbound

This paper cites Anygpt: Unified multimodal llm with discrete sequence modeling, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Anygpt: Unified multimodal llm with discrete sequence modeling, 2025

Reference 43

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fcab27b-cfc4-4226-84c3-cffd89c4de0b · outbound

This paper cites A-vl: Adaptive attention for large vision- language models, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training A-vl: Adaptive attention for large vision- language models, 2025

Reference 44

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3202d961-ff2e-4165-b8cb-f45f634bbcd5 · outbound

This paper cites Llava-mini: Efficient image and video large multimodal models with one vision token, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Llava-mini: Efficient image and video large multimodal models with one vision token, 2025

Reference 45

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 24525677-03b9-475a-a718-cc04d98432cd · outbound

This paper cites Himix: Reducing computational com- plexity in large vision-language models, 2025.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Himix: Reducing computational com- plexity in large vision-language models, 2025

Reference 46

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce075547-819f-487b-a606-2e0bc49fa037 · outbound

This paper cites Ar- gus: A compact and versatile foundation model for vision.

On the Limits of Token Reduction for Efficient Unified Vision Language Training Ar- gus: A compact and versatile foundation model for vision

Reference 47

Resolution: unresolved; no resolver link, observed 2026-06-28T16:57:56.333595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.