Paper Citation Record · LEDGER

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

As of 14 August 2026, this Paper Citation Record lists 59 of 59 outbound references and 5 inbound Pith citation observations for arXiv:2411.14228.

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

pith.paper-citation-record.v1
2411.14228 v1

Coverage vector

measured 59 of 59 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T15:28:37.837338Z

measured 64 of 64 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+00:00

measured 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-11T17:38:56.772419Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-12T07:56:35.564771Z

Reference resolution

59 of 59 outbound references displayed

verified exact0
verified fuzzy22
unresolved37
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 85981306-6ba3-47e3-962a-d3eab24d426f · outbound

This paper cites Menick, Sebastian Borgeaud, Andy Brock, Aida Nematzadeh, Sa- hand Sharifzadeh, Mikolaj Binkowski, Ricardo Barreira, Oriol Vinyals, Andrew Zisserman, and Kar ´en Simonyan.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Menick, Sebastian Borgeaud, Andy Brock, Aida Nematzadeh, Sa- hand Sharifzadeh, Mikolaj Binkowski, Ricardo Barreira, Oriol Vinyals, Andrew Zisserman, and Kar ´en Simonyan

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.877324Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation be6b8822-90b2-4eb1-910c-1e857f7702bb · outbound

This paper cites HiRED: Attention-Guided Token Dropping for Efficient Inference of High-Resolution Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression HiRED: Attention-Guided Token Dropping for Efficient Inference of High-Resolution Vision-Language Models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.002477Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44541421-4f6f-428c-b44a-f559bce6cead · outbound

This paper cites Qwen-vl: A versatile vision-language model for un- derstanding, localization, text reading, and beyond.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Qwen-vl: A versatile vision-language model for un- derstanding, localization, text reading, and beyond

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.720353Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 84fcb041-986c-48ad-bba5-6b218014217a · outbound

This paper cites Introducing our multimodal models, 2023.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Introducing our multimodal models, 2023

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.597815Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f18d397b-de95-4bc2-ae3b-3ab8387e1d59 · outbound

This paper cites Matryoshka Multimodal Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Matryoshka Multimodal Models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.016869Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6d6801c9-0478-49c7-9347-393dabd23a54 · outbound

This paper cites Madtp: Multi- modal alignment-guided dynamic token pruning for accel- erating vision-language transformer.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Madtp: Multi- modal alignment-guided dynamic token pruning for accel- erating vision-language transformer

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.021723Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e88b20a6-2d67-49ee-8bee-3bed8fda4fbc · outbound

This paper cites ALLaVA: Harnessing GPT4V-Synthesized Data for Lite Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression ALLaVA: Harnessing GPT4V-Synthesized Data for Lite Vision-Language Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.026326Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97077036-845c-4526-8562-84f99b63e1bc · outbound

This paper cites Allava: Harness- ing gpt4v-synthesized data for a lite vision-language model,.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Allava: Harness- ing gpt4v-synthesized data for a lite vision-language model,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.570415Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 9521270f-9bc6-4c7b-8053-beeedba0edeb · outbound

This paper cites Geoqa: A geomet- ric question answering benchmark towards multimodal nu- merical reasoning.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Geoqa: A geomet- ric question answering benchmark towards multimodal nu- merical reasoning

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.553874Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 6bdbb30c-4332-4629-becb-aa9a8773fa17 · outbound

This paper cites Dragonfly: Multi-Resolution Zoom-In Encoding Enhances Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Dragonfly: Multi-Resolution Zoom-In Encoding Enhances Vision-Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.040643Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7212cb04-4fd8-42ff-91c5-b22068d28a7d · outbound

This paper cites Open-llava-next: An open- source implementation of llava-next series for facilitating the large multi-modal model community.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Open-llava-next: An open- source implementation of llava-next series for facilitating the large multi-modal model community

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.537965Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation b86bb992-3cc1-44cd-b325-9dec4103f217 · outbound

This paper cites ShareGPT4V: Improving Large Multi-Modal Models with Better Captions.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression ShareGPT4V: Improving Large Multi-Modal Models with Better Captions

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.050476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2204a2cf-eeb2-4123-b637-d03e7f884779 · outbound

This paper cites An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.075718Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c15604c6-b00f-4649-8317-86262a8f22bb · outbound

This paper cites an unresolved cited work.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-12T15:28:39.520715Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 6d717807-1f12-4c24-bb31-d9a95e95ffa3 · outbound

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

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression An image is worth 16x16 words: Transformers for image recognition at scale

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.156258Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7c1f498-9289-42b5-99ba-9d5d34673e90 · outbound

This paper cites Switch transformers: Scaling to trillion parameter models with sim- ple and efficient sparsity.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Switch transformers: Scaling to trillion parameter models with sim- ple and efficient sparsity

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.381261Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8cde052b-b0ad-451d-9de7-31cae2645616 · outbound

This paper cites MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.209797Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8c4dcd8-496a-4d67-bf48-0cea65accaed · outbound

This paper cites Matryoshka Query Transformer for Large Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Matryoshka Query Transformer for Large Vision-Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.214098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7f679ab-04fa-4717-bd8a-835a9f7d7778 · outbound

This paper cites Bliva: A simple multimodal llm for better handling of text-rich visual questions.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Bliva: A simple multimodal llm for better handling of text-rich visual questions

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.218265Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aa5d41bb-cc54-416c-bcc9-229d8d97e611 · outbound

This paper cites Hudson and Christopher D.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Hudson and Christopher D

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.326081Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 835b0f59-ef62-4db7-89d3-326a2d250049 · outbound

This paper cites Dvqa: Understanding data visualizations via ques- tion answering.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Dvqa: Understanding data visualizations via ques- tion answering

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.311235Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c65a56df-4e85-4a4d-83f2-ccd473697ed5 · outbound

This paper cites A diagram is worth a dozen images.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression A diagram is worth a dozen images

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.296116Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 4da35707-260f-41f2-b308-7ea4c9c3294b · outbound

This paper cites Ocr-free document understanding transformer.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Ocr-free document understanding transformer

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.281655Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 07c3197e-4bb8-41db-a3ee-5068bb653d2e · outbound

This paper cites OtterHD: A High-Resolution Multi-modality Model.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression OtterHD: A High-Resolution Multi-modality Model

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.243021Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 496fa244-7501-45eb-97aa-517fc9a64638 · outbound

This paper cites an unresolved cited work.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-08-12T15:28:39.198504Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation b5fd66df-daab-4f6a-9bca-a0aaa5b13eec · outbound

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

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Evaluating object hallucination in large vision-language models

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.151575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 10f5cb9f-cab3-4929-912b-d7b3d1096199 · outbound

This paper cites Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.258045Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7bb0d8d-92da-486f-83a8-aea9522449da · outbound

This paper cites Mon- key: Image resolution and text label are important things for large multi-modal models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Mon- key: Image resolution and text label are important things for large multi-modal models

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.137689Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 949bee26-8fbd-4dbb-808c-66e5fff87b8e · outbound

This paper cites MoE-LLaVA: Mixture of Experts for Large Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression MoE-LLaVA: Mixture of Experts for Large Vision-Language Models

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.269526Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 83a27c8e-2a75-4482-b212-2e70bf6e2fe5 · outbound

This paper cites Open-llava-next: An open- source implementation of llava-next series for facilitating the large multi-modal model community.

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.123104Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation b41d26f8-b5bd-4fe4-8482-4facd44c33fb · outbound

This paper cites SPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression SPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.279341Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b2d0748-67b0-40b2-8e29-7a8d8f00469a · outbound

This paper cites Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.284569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c24928d-5701-4377-a67b-653a8e538ef6 · outbound

This paper cites Visual instruction tuning.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Visual instruction tuning

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:39.054195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d4301e7b-0e88-45d2-bf19-4778c07ab290 · outbound

This paper cites Improved baselines with visual instruction tuning.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Improved baselines with visual instruction tuning

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.903964Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation fc93e0bd-653e-4f05-a1ff-96bfb8ef4c90 · outbound

This paper cites Llava-next: Im- proved reasoning, ocr, and world knowledge, 2024.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Llava-next: Im- proved reasoning, ocr, and world knowledge, 2024

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.400560Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f907f837-daa6-4e16-a89b-b5a628eb484e · outbound

This paper cites MMBench: Is Your Multi-modal Model an All-around Player?.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression MMBench: Is Your Multi-modal Model an All-around Player?

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.424539Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2086919b-ee04-48d5-a617-3364ef15fbc9 · outbound

This paper cites TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.430152Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e31518c-808a-4dbb-be17-301aa00dfb90 · outbound

This paper cites Learn to explain: Multimodal reasoning via thought chains for science question answering.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Learn to explain: Multimodal reasoning via thought chains for science question answering

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.876420Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 266936f7-9d45-42fd-bf87-d890353fd29a · outbound

This paper cites Feast Your Eyes: Mixture-of-Resolution Adaptation for Multimodal Large Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Feast Your Eyes: Mixture-of-Resolution Adaptation for Multimodal Large Language Models

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.439399Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e30d2dee-8f06-461a-af90-96c6184f7bfa · outbound

This paper cites Joty, and Enamul Hoque.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Joty, and Enamul Hoque

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.444054Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8e8dca30-61ce-4018-a297-bb4ecd828cc2 · outbound

This paper cites an unresolved cited work.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Unresolved cited work

Reference 41

Resolution: unresolved; raw_fallback, observed 2026-08-12T15:28:38.732723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d5548868-4c7d-49a8-a8f9-b94fbeb9a2a0 · outbound

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

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Learning transferable visual models from natural language supervision

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.457732Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ac31051f-4420-4236-b7dd-9be4eef286ba · outbound

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

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Llava-prumerge: Adaptive token reduction for efficient large multimodal models

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.462581Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3b18de5-609a-4227-9cdf-fd753c2f34c9 · outbound

This paper cites Towards VQA models that can read.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Towards VQA models that can read

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.621897Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 9663235a-df3b-437f-95f5-446efddfc313 · outbound

This paper cites Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.471245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0735f891-3805-45f3-9512-989adea90ff9 · outbound

This paper cites Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.476018Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68937910-ccfd-4f15-831d-2fa37fd7d194 · outbound

This paper cites Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.480933Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation adf3e0c0-7b68-4965-945b-5221e8622735 · outbound

This paper cites LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.486326Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68cb61a9-185a-4ec7-b68b-010ddd4ca265 · outbound

This paper cites mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.491494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6ce1aae-48ec-4cef-8db7-c1a9c615a421 · outbound

This paper cites Mm-vet: Evaluating large multimodal models for integrated capabilities.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Mm-vet: Evaluating large multimodal models for integrated capabilities

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.573491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c4ce4de7-7a45-411c-a2a9-754906917c48 · outbound

This paper cites TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.501180Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fac41a21-f0be-49ee-aa91-a5f5766ced3d · outbound

This paper cites TextHawk2: A Large Vision-Language Model Excels in Bilingual OCR and Grounding with 16x Fewer Tokens.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression TextHawk2: A Large Vision-Language Model Excels in Bilingual OCR and Grounding with 16x Fewer Tokens

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.529815Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1776f1c2-6979-4d83-a4ee-428e21b33d76 · outbound

This paper cites Token-level Correlation-guided Compression for Efficient Multimodal Document Understanding.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Token-level Correlation-guided Compression for Efficient Multimodal Document Understanding

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.620931Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c85c23ed-b540-42fa-b5ae-eb93a40efafd · outbound

This paper cites Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-12T15:28:37.702245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1009fd42-b04b-4819-bfc0-7c95510b4de7 · outbound

This paper cites This is used to illustrate that a learned metric rather than hand-crafted will solve the prob- lem of performance reduction.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression This is used to illustrate that a learned metric rather than hand-crafted will solve the prob- lem of performance reduction

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.558449Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 65663ce6-e6ef-43a3-8c39-0f7064f82c4e · outbound

This paper cites 6 and Fig.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression 6 and Fig

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.542583Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e2054ada-94f5-42f5-b2fd-1bb84204a22b · outbound

This paper cites an unresolved cited work.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Unresolved cited work

Reference 58

Resolution: unresolved; raw_fallback, observed 2026-08-12T15:28:38.524062Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 81b7ee5d-849d-44fa-bec4-24c0724dfa62 · outbound

This paper cites During our training process, under the constraint of balance loss, the model is required to select three different visual scales with as equal probabil- ity as possible.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression During our training process, under the constraint of balance loss, the model is required to select three different visual scales with as equal probabil- ity as possible

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T15:28:38.426533Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 0ba032dd-de4c-42e2-a7e7-fd035bfe3871 · outbound

This paper cites an unresolved cited work.

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression Unresolved cited work

Reference 2279

Resolution: unresolved; raw_fallback, observed 2026-08-12T15:28:38.851506Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Pith citing papers

Observation 81522011-0c70-4c50-b4e6-f46a7adcdf3e · inbound

LLaVA-Zip: Adaptive Visual Token Compression with Intrinsic Image Information cites this paper.

LLaVA-Zip: Adaptive Visual Token Compression with Intrinsic Image Information FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T17:38:56.772419Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01988274-4b3b-42ef-bd1b-1caf36fbd444 · inbound

METEOR: Multi-Encoder Collaborative Token Pruning for Efficient Vision Language Models cites this paper.

METEOR: Multi-Encoder Collaborative Token Pruning for Efficient Vision Language Models FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-06T13:18:06.528428Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dfa3f6f5-d21d-4709-aecc-aeed2acded77 · inbound

An Efficient Token Compression Framework for Visual Object Tracking cites this paper.

An Efficient Token Compression Framework for Visual Object Tracking FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

Reference 73

Resolution: verified exact; arxiv_id, observed 2026-05-12T07:56:35.600359Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 924ecc97-c796-401d-83cf-d8baec19dfba · inbound

ET-Prune: Evidence-Aware Dynamic Budgeting for Visual Token Pruning in Text-Rich MLLMs cites this paper.

ET-Prune: Evidence-Aware Dynamic Budgeting for Visual Token Pruning in Text-Rich MLLMs FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-04T17:23:16.296228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4bfe39b-061b-4146-86c4-abc0b20949bd · inbound

Learning to Predict Middle-Layer Attention in MLLMs for Visual Token Prunin cites this paper.

Learning to Predict Middle-Layer Attention in MLLMs for Visual Token Prunin FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-10T04:30:14.375574Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.