Paper Citation Record · LEDGER

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

As of 18 August 2026, this Paper Citation Record lists 100 of 102 outbound references and 10 inbound Pith citation observations for arXiv:2412.03248.

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

pith.paper-citation-record.v1
2412.03248 v2

Coverage vector

measured 100 of 102 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T22:42:35.606021Z

measured 110 of 110 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 10 of 10 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-16T11:46:40.890317Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T10:41:03.999598Z

Reference resolution

100 of 102 outbound references displayed

verified exact0
verified fuzzy51
unresolved49
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation b0029a0e-2d85-4f64-bf21-3b59605dedd3 · outbound

This paper cites Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.441164Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a517098-19be-47ea-967d-9a3c809b86a4 · outbound

This paper cites Conditional Computation in Neural Networks for faster models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Conditional Computation in Neural Networks for faster models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.450087Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66289086-3f93-4906-8a59-72f723f4ab3c · outbound

This paper cites Token merging: Your ViT but faster.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Token merging: Your ViT but faster

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.457746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fd99e16d-42e9-4d6b-8619-f97814eb7df5 · outbound

This paper cites Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.466118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f07c04b-41a2-471e-a262-8a5c6b562efb · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning An image is worth 1/2 tokens after layer 2: Plug-and-play inference accelera- tion for large vision-language models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.474254Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97c3365e-a694-4982-956c-8e52125de695 · outbound

This paper cites Gonzalez, Ion Stoica, and Eric P.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Gonzalez, Ion Stoica, and Eric P

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.492269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d6522194-ca2d-4132-b55a-6df2d3c786b4 · outbound

This paper cites Generating Long Sequences with Sparse Transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Generating Long Sequences with Sparse Transformers

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.508061Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbc93ba2-a3bf-4978-8744-ed063811f16d · outbound

This paper cites InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.522544Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 72f6d86b-a88b-4be4-a8fb-fc4c955b7f61 · outbound

This paper cites FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.531047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e36d8d08-4439-4c6c-8880-2dfe9a25e92c · outbound

This paper cites an unresolved cited work.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Unresolved cited work

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.537975Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4bca0f33-9232-44b1-8c62-abfb515c7a90 · outbound

This paper cites HeatViT: Hardware-Efficient Adap- tive Token Pruning for Vision Transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning HeatViT: Hardware-Efficient Adap- tive Token Pruning for Vision Transformers

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.547798Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b394b5fd-0e4c-410a-83ad-899aa9c32586 · outbound

This paper cites Glam: Efficient scaling of language models with mixture-of-experts.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Glam: Efficient scaling of language models with mixture-of-experts

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.559597Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be2ceb51-bacf-459b-9560-690a844a44ab · outbound

This paper cites Hsu, and Shang-Hong Lai.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Hsu, and Shang-Hong Lai

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.568230Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c4a36c6-2fe0-4eeb-b36b-2734fda1b327 · outbound

This paper cites Adaptive Token Sampling for Efficient Vision Transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Adaptive Token Sampling for Efficient Vision Transformers

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.628752Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b902260-180f-45a8-9d41-51803814a93a · outbound

This paper cites Spatially adaptive computation time for residual networks.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Spatially adaptive computation time for residual networks

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.720208Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37316ee7-8a7b-4b70-8305-4fb2a3363bd7 · outbound

This paper cites Mme: A comprehensive evaluation benchmark for multimodal large language models, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mme: A comprehensive evaluation benchmark for multimodal large language models, 2024

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.813015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94e89941-f4ea-47ea-abe5-7ebd81815d09 · outbound

This paper cites Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:32.928121Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e51cbe8-edf1-435c-96f1-073d904d239d · outbound

This paper cites PoWER-BERT: Accelerating BERT Inference via Progressive Word-Vector Elimination.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning PoWER-BERT: Accelerating BERT Inference via Progressive Word-Vector Elimination

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.027600Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 35e73f7a-154d-4448-bfa4-a893656e4d6b · outbound

This paper cites Making the v in vqa matter: El- evating the role of image understanding in visual question answering, 2017.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Making the v in vqa matter: El- evating the role of image understanding in visual question answering, 2017

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.040285Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f38655c9-030c-4e82-a64c-71a71c8d5e58 · outbound

This paper cites Speedboost: Anytime pre- diction with uniform near-optimality.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Speedboost: Anytime pre- diction with uniform near-optimality

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.048908Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be8f9fa3-87c4-4a3b-96ef-f32f91eac3f2 · outbound

This paper cites Mamba: Linear-time sequence modeling with selective state spaces, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mamba: Linear-time sequence modeling with selective state spaces, 2024

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.054914Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14e87409-3aef-4da1-93ee-bcf6254b8109 · outbound

This paper cites Dynamic neural networks: A sur- vey.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Dynamic neural networks: A sur- vey

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.060957Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df0600ee-3fd0-4408-9720-d7210b29bac3 · outbound

This paper cites Learning anytime predictions in neural net- works via adaptive loss balancing.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Learning anytime predictions in neural net- works via adaptive loss balancing

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.067491Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14db3d63-a0d8-47ed-932d-dde1e2027987 · outbound

This paper cites Longrecipe: Recipe for efficient long context generalization in large language models, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Longrecipe: Recipe for efficient long context generalization in large language models, 2024

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.079353Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 31e093b3-e80b-421d-8892-b1679769d893 · outbound

This paper cites Language is not all you need: Aligning perception with language mod- els.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Language is not all you need: Aligning perception with language mod- els

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.088729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 362ebd48-17f1-4b06-8dc4-1fd2086584d4 · outbound

This paper cites Gqa: A new dataset for real-world visual reasoning and compositional question answering.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Gqa: A new dataset for real-world visual reasoning and compositional question answering

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.096469Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14874ec9-b7bb-47fb-8efa-37e38cf72d37 · outbound

This paper cites Anytime recognition with routing convolutional networks.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Anytime recognition with routing convolutional networks

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.101574Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2bf2311f-1c4e-4b38-a586-8aefa9dec776 · outbound

This paper cites Anytime recognition of objects and scenes.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Anytime recognition of objects and scenes

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.107004Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7c55cd4-6d31-4ac9-8cb4-4f394139b434 · outbound

This paper cites Length-Adaptive Transformer: Train Once with Length Drop, Use Anytime with Search.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Length-Adaptive Transformer: Train Once with Length Drop, Use Anytime with Search

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.113003Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8500cadb-5d0b-45a9-b01d-ff4130395d82 · outbound

This paper cites Learned Token Pruning for Transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Learned Token Pruning for Transformers

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.119961Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b730ae6a-7ada-49ba-8c56-6820884d089d · outbound

This paper cites SPViT: Enabling Faster Vision Transform- ers Latency-Aware Soft Token Pruning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning SPViT: Enabling Faster Vision Transform- ers Latency-Aware Soft Token Pruning

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:40.139405Z

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 d118ceed-a1ec-48ea-9f7c-790a8a5b2c11 · outbound

This paper cites Text-conditioned resampler for long form video understanding, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Text-conditioned resampler for long form video understanding, 2024

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:40.097815Z

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 49b89833-6f25-43fd-88c9-fb5def88c31e · outbound

This paper cites LLaVA-OneVision: Easy Visual Task Transfer.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning LLaVA-OneVision: Easy Visual Task Transfer

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.145410Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5837c2c-d2bc-42ad-b801-f47ec19ea465 · outbound

This paper cites 2d or not 2d? adaptive 3d convolution selec- tion for efficient video recognition.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning 2d or not 2d? adaptive 3d convolution selec- tion for efficient video recognition

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:40.044840Z

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 240bd5a9-c36f-4a23-9933-a287b6d98e0f · outbound

This paper cites BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.163460Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 450e9e6a-6069-4479-a81b-3a6d5b58cd4c · outbound

This paper cites Mvbench: A comprehensive multi- modal video understanding benchmark, 2023.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mvbench: A comprehensive multi- modal video understanding benchmark, 2023

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.992675Z

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 880428ec-814b-47b5-978f-618641a884b9 · outbound

This paper cites Mvbench: A comprehensive multi- modal video understanding benchmark, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mvbench: A comprehensive multi- modal video understanding benchmark, 2024

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.915018Z

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 fa7bce63-e9cf-4fab-bb1a-08ed492d2570 · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Evaluating object hallucination in large vision-language models, 2023

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.830333Z

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 03390d55-db6f-45ac-b385-58661ceaa3a4 · outbound

This paper cites Video-LLaVA: Learning United Visual Representation by Alignment Before Projection.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.499475Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c29085ce-8130-491a-9e3e-94c02e51fc3a · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Boosting multimodal large language models with visual to- kens withdrawal for rapid inference

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.710130Z

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 6b146270-361b-468b-95f7-99f9db001ecc · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Improved baselines with visual instruction tuning, 2023

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.611242Z

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 040a0f2b-e501-42b1-85f5-12aa6a7fb64c · outbound

This paper cites Adaptive Sparse ViT: Towards Learnable Adaptive Token Pruning by Fully Exploiting Self-Attention.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Adaptive Sparse ViT: Towards Learnable Adaptive Token Pruning by Fully Exploiting Self-Attention

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.536591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2cd49640-e815-4fd6-9613-86b83b60705e · outbound

This paper cites Mm- bench: Is your multi-modal model an all-around player?,.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mm- bench: Is your multi-modal model an all-around player?,

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.524869Z

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 8258f6ad-9655-464a-9a08-a25085257104 · outbound

This paper cites Learn to explain: Multimodal reasoning via thought chains for science question answering, 2022.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Learn to explain: Multimodal reasoning via thought chains for science question answering, 2022

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.488095Z

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 7005e5a2-b4fd-4291-996b-c573d4e492d6 · outbound

This paper cites Video-chatgpt: Towards detailed video understanding via large vision and language models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Video-chatgpt: Towards detailed video understanding via large vision and language models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.565798Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86a0b85c-9408-4b25-b55b-2cf851f878f8 · outbound

This paper cites Egoschema: A diagnostic benchmark for very long- form video language understanding, 2023.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Egoschema: A diagnostic benchmark for very long- form video language understanding, 2023

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.384758Z

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 d8e3a40c-93ca-4ac9-8563-51d4dc79cede · outbound

This paper cites Llama3.2.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Llama3.2

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.235114Z

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 457fb89c-8351-4a77-b0ad-bbda2d801f59 · outbound

This paper cites Adavit: Adaptive vision transformers for efficient image recogni- tion.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Adavit: Adaptive vision transformers for efficient image recogni- tion

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:39.061038Z

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 d18a5a18-d1d9-49d0-bd2d-9e07a68b1f16 · outbound

This paper cites Ar-net: Adaptive frame resolution for effi- cient action recognition.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Ar-net: Adaptive frame resolution for effi- cient action recognition

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.958217Z

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 4ac4217a-9349-4990-8340-b75bfc0a5d3f · outbound

This paper cites Compositional chain of thought prompting for large multimodal models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Compositional chain of thought prompting for large multimodal models

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.887809Z

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 5c6b7c86-c7ae-4b2c-8ef6-895367126778 · outbound

This paper cites Encod- ing and controlling global semantics for long-form video question answering, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Encod- ing and controlling global semantics for long-form video question answering, 2024

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.804751Z

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 9a3428c4-d063-4309-b640-4bb0824975d0 · outbound

This paper cites an unresolved cited work.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Unresolved cited work

Reference 52

Resolution: unresolved; raw_fallback, observed 2026-08-11T22:42:38.754751Z

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 d97414c5-bacb-4d04-b009-efc026c23638 · outbound

This paper cites Gpt-4 technical report, 2023.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Gpt-4 technical report, 2023

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:33.655655Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff77bf7b-c85c-4bfa-a91f-64abccbfbcf1 · outbound

This paper cites Training lan- guage models to follow instructions with human feedback.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Training lan- guage models to follow instructions with human feedback

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.590187Z

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 dd422b71-84c5-4a32-aaa8-3cf4bbcc2929 · outbound

This paper cites The pagerank citation ranking: Bringing order to the web.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning The pagerank citation ranking: Bringing order to the web

Reference 55

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.507077Z

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 0e6327bc-7b09-4b33-8b10-550cec56181b · outbound

This paper cites Ia-red2: Interpretability-aware redundancy reduction for vision transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Ia-red2: Interpretability-aware redundancy reduction for vision transformers

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.454239Z

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 f83faf88-c2f2-4738-ad7d-08674c4912e0 · outbound

This paper cites Perception test: A diagnostic benchmark for multimodal video models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Perception test: A diagnostic benchmark for multimodal video models

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.428599Z

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 5accbd04-bbae-4721-9691-191243fa9763 · outbound

This paper cites Learning transferable visual models from natural language supervi- sion.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Learning transferable visual models from natural language supervi- sion

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.380154Z

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 1a0956f3-402a-4f0d-a575-423db1e20a99 · outbound

This paper cites DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.333199Z

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 7c80718b-24b9-47fe-b54e-1dbce6f47b46 · outbound

This paper cites Finding the sweet spot: Analysis and improve- ment of adaptive inference in low resource settings.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Finding the sweet spot: Analysis and improve- ment of adaptive inference in low resource settings

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.309827Z

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 ec62986d-32f0-4ecf-8b81-fb2b003c378f · outbound

This paper cites Interpolating video-llms: Toward longer- sequence lmms in a training-free manner, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Interpolating video-llms: Toward longer- sequence lmms in a training-free manner, 2024

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.284187Z

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 1edc9274-79e0-4e84-911f-923a55456b2f · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Llava-prumerge: Adaptive token reduction for efficient large multimodal models

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.201277Z

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 6ede8a3c-8eca-4565-a815-9aaf4d00261e · outbound

This paper cites Towards vqa models that can read, 2019.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Towards vqa models that can read, 2019

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.115737Z

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 0b4c81a8-7082-49f3-b23d-8aed28fd2fb5 · outbound

This paper cites Dynamic Token Pruning in Plain Vision Trans- formers for Semantic Segmentation.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Dynamic Token Pruning in Plain Vision Trans- formers for Semantic Segmentation

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:38.073037Z

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 1d22d025-eb75-4db5-886f-97daac3cb8c9 · outbound

This paper cites Dycoke: Dynamic compression of tokens for fast video large language models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Dycoke: Dynamic compression of tokens for fast video large language models

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.973209Z

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 038999bd-7622-49e7-b053-8cc33e73319c · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning LLaMA: Open and Efficient Foundation Language Models

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.245840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0789ecf3-c8ce-494a-95d9-7643ad399a1d · outbound

This paper cites SpAtten: Ef- ficient Sparse Attention Architecture with Cascade Token and Head Pruning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning SpAtten: Ef- ficient Sparse Attention Architecture with Cascade Token and Head Pruning

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.945733Z

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 47e67a27-d0a6-4fcb-a93f-35500c172fb7 · outbound

This paper cites Zero- tprune: Zero-shot token pruning through leveraging of the attention graph in pre-trained transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Zero- tprune: Zero-shot token pruning through leveraging of the attention graph in pre-trained transformers

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.918130Z

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 1227f3ed-7a15-4a42-9eee-8cf8e7d7434e · outbound

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

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.320401Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e4b5ed8-9e99-4eab-9beb-bbdf5ef19fd1 · outbound

This paper cites Skipnet: Learning dynamic routing in convolutional networks.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Skipnet: Learning dynamic routing in convolutional networks

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.882126Z

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 0e2bba78-430d-4540-965f-c24aea22fe1a · outbound

This paper cites Longllava: Scaling multi-modal llms to 1000 images efficiently via a hybrid architecture, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Longllava: Scaling multi-modal llms to 1000 images efficiently via a hybrid architecture, 2024

Reference 71

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.836613Z

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 f5a96be5-5d41-47b8-9b50-2bcdbbba7819 · outbound

This paper cites Not all images are worth 16x16 words: Dynamic transformers for efficient image recognition.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Not all images are worth 16x16 words: Dynamic transformers for efficient image recognition

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.792671Z

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 c638345b-7acb-4397-a2b1-5bfe2aea6aa0 · outbound

This paper cites Videollamb: Long-context video understanding with recur- rent memory bridges, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Videollamb: Long-context video understanding with recur- rent memory bridges, 2024

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.768762Z

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 a4c1ce93-90a4-4ac0-802e-7c8f7fd582cd · outbound

This paper cites Visual context win- dow extension: A new perspective for long video under- standing, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Visual context win- dow extension: A new perspective for long video under- standing, 2024

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.744662Z

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 23241af9-9f4e-43c9-9542-b371537fe413 · outbound

This paper cites Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.679928Z

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 cfcc0320-87b6-4b8f-8565-bd8a1bdad02c · outbound

This paper cites Longvlm: Efficient long video under- standing via large language models, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Longvlm: Efficient long video under- standing via large language models, 2024

Reference 76

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.656283Z

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 47846cea-a78b-4d35-8e4d-85e9b8e37c97 · outbound

This paper cites Blockdrop: Dynamic inference paths in residual net- works.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Blockdrop: Dynamic inference paths in residual net- works

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.624531Z

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 4648a01f-e97b-44bc-b8d3-aa462314bc7a · outbound

This paper cites Next-qa:next phase of question-answering to explaining temporal actions, 2021.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Next-qa:next phase of question-answering to explaining temporal actions, 2021

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.570100Z

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 69b0ac4a-7df3-4171-a32e-707d0c6a62bf · outbound

This paper cites Conical visual concen- tration for efficient large vision-language models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Conical visual concen- tration for efficient large vision-language models

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.531799Z

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 e2cfcd85-85d0-4b5c-a3e2-d397f22d1590 · outbound

This paper cites PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.808922Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c679c919-d041-4e74-ad92-17cc044aa0bf · outbound

This paper cites Smartadapt: Multi-branch object detection framework for videos on mo- biles.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Smartadapt: Multi-branch object detection framework for videos on mo- biles

Reference 81

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.454749Z

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 2e149b68-1f3c-4aba-b4ea-299c66d8b911 · outbound

This paper cites The greedy miser: learning under test-time budgets.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning The greedy miser: learning under test-time budgets

Reference 82

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.386268Z

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 2b42e201-f2c7-4aa2-a3ea-52167ea5b291 · outbound

This paper cites Learning to inference adaptively for multimodal large language models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Learning to inference adaptively for multimodal large language models

Reference 83

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.337698Z

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 071fcb59-3f8e-4437-acd7-d53341c590dc · outbound

This paper cites LongVILA: Scaling Long-Context Visual Language Models for Long Videos.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning LongVILA: Scaling Long-Context Visual Language Models for Long Videos

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.850695Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2a1cada-6559-4c95-85e3-a887118f5636 · outbound

This paper cites Qwen2 Technical Report.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Qwen2 Technical Report

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.856995Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 33c82efb-326a-4499-83ab-a3caa6c57001 · outbound

This paper cites TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:34.872462Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a7689da-4d9d-4c25-8847-cc9589d1b9b4 · outbound

This paper cites mplug-owl3: Towards long image-sequence understanding in multi-modal large language models, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning mplug-owl3: Towards long image-sequence understanding in multi-modal large language models, 2024

Reference 87

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.274750Z

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 d628cffe-e20e-45bf-ad9f-5f37ede00a6d · outbound

This paper cites Fit and prune: Fast and training-free visual token pruning for multi-modal large language models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Fit and prune: Fast and training-free visual token pruning for multi-modal large language models

Reference 88

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.208625Z

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 2b0bff02-7148-49f2-b7bc-6a5c1ea8452b · outbound

This paper cites Alvarez, Arun Mallya, Jan Kautz, and Pavlo Molchanov.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Alvarez, Arun Mallya, Jan Kautz, and Pavlo Molchanov

Reference 89

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.163076Z

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 0f37e1d6-9c72-451e-9a23-fd816d9d0a07 · outbound

This paper cites Llm inference unveiled: Survey and roofline model in- sights, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Llm inference unveiled: Survey and roofline model in- sights, 2024

Reference 90

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.130770Z

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 9c8afaca-8878-42c6-9f66-921e155a77d1 · outbound

This paper cites Sigmoid loss for language image pre-training.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Sigmoid loss for language image pre-training

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.411927Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30ce5201-7e5d-4032-bfee-955a1cec7fbe · outbound

This paper cites Long Context Transfer from Language to Vision.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Long Context Transfer from Language to Vision

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.454216Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3cc18f74-ef5e-440c-8bf0-04d87bcdaf89 · outbound

This paper cites OPT: Open Pre-trained Transformer Language Models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning OPT: Open Pre-trained Transformer Language Models

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.470309Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf704827-ccb7-4c18-976d-ddc899f90bf2 · outbound

This paper cites Llava-next: A strong zero-shot video understanding model,.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Llava-next: A strong zero-shot video understanding model,

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.480309Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8c2bfff-7ab0-49dd-a3b8-863289b432fc · outbound

This paper cites Video instruction tuning with synthetic data, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Video instruction tuning with synthetic data, 2024

Reference 95

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:37.049515Z

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 e38301fb-a2b6-444d-83cd-7ae159be5830 · outbound

This paper cites Sparsevlm: Vi- sual token sparsification for efficient vision-language model inference.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Sparsevlm: Vi- sual token sparsification for efficient vision-language model inference

Reference 96

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:36.961412Z

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 f041d933-1902-412a-9ac3-1acf2b510799 · outbound

This paper cites Multimodal Chain-of-Thought Reasoning in Language Models.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Multimodal Chain-of-Thought Reasoning in Language Models

Reference 97

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.510876Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e68b2d4-546a-4dd0-8c91-07175928aee8 · outbound

This paper cites Treat Visual Tokens as Text? But Your MLLM Only Needs Fewer Efforts to See.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Treat Visual Tokens as Text? But Your MLLM Only Needs Fewer Efforts to See

Reference 98

Resolution: unresolved; no resolver link, observed 2026-08-11T22:42:35.570387Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ae94114-09fb-465c-b2fb-d5ca5bed3427 · outbound

This paper cites Beyond embeddings: The promise of visual table in vi- sual reasoning.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Beyond embeddings: The promise of visual table in vi- sual reasoning

Reference 99

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:36.926701Z

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 a9a29163-bbce-4992-842b-e601b02006ff · outbound

This paper cites Mlvu: A comprehensive benchmark for multi- task long video understanding, 2024.

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Mlvu: A comprehensive benchmark for multi- task long video understanding, 2024

Reference 100

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T22:42:36.681732Z

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

Observation 7cf09e0b-8c67-4d98-8be8-2fb881ea285f · inbound

A Survey on Large Language Model Acceleration based on KV Cache Management cites this paper.

A Survey on Large Language Model Acceleration based on KV Cache Management AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 109

Resolution: unresolved; no resolver link, observed 2026-08-11T00:38:47.668642Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ebd6af17-6836-4c4e-bfc1-b15169f0b67c · inbound

FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models cites this paper.

FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-10T23:09:25.185790Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 77e2f337-f90d-448d-b986-cf87959df91b · inbound

Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark cites this paper.

Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 174

Resolution: unresolved; no resolver link, observed 2026-08-16T11:46:40.890317Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f61d0b3a-33cc-4d09-9490-858863c934ec · inbound

Streamline Without Sacrifice -- Squeeze out Computation Redundancy in LMM cites this paper.

Streamline Without Sacrifice -- Squeeze out Computation Redundancy in LMM AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T15:15:12.434364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ed619fe-2182-414b-afff-bd1a1ca8dd69 · inbound

AdaTP: Attention-Debiased Token Pruning for Video Large Language Models cites this paper.

AdaTP: Attention-Debiased Token Pruning for Video Large Language Models AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-07T14:08:10.915558Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b344fd13-b7d1-46a2-af72-80b526c10748 · inbound

Fast3D: Accelerating 3D Multi-modal Large Language Models for Efficient 3D Scene Understanding cites this paper.

Fast3D: Accelerating 3D Multi-modal Large Language Models for Efficient 3D Scene Understanding AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-06T18:03:03.555597Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d65a803-0200-4e6b-aa91-044e0e65175b · 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 AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-06T13:18:06.370053Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6074cee1-1481-471e-9de5-4e6c03422122 · inbound

Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects cites this paper.

Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 13

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T23:45:50.915833Z

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 1228267b-abf2-44b3-99d4-d9c23b22df29 · inbound

POINTS-Long: Adaptive Dual-Mode Visual Reasoning in MLLMs cites this paper.

POINTS-Long: Adaptive Dual-Mode Visual Reasoning in MLLMs AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 122

Resolution: verified exact; arxiv_id, observed 2026-05-11T10:41:04.005982Z

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 6db44e5d-2150-469d-81d6-7bd821c2662a · inbound

Stable Curves, Unstable Items: Item-Level Scaling Heterogeneity in Video LLMs cites this paper.

Stable Curves, Unstable Items: Item-Level Scaling Heterogeneity in Video LLMs AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-10T16:42:44.443670Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.