TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs | Paper Citation Record

Observation 0bf742f8-da7f-42ec-84c5-bb0a48f8177d · outbound

This paper cites Vision-language models struggle to align entities across modalities.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Vision-language models struggle to align entities across modalities

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.224783Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a0966a1-e105-4bf6-838f-4d5fa7fe0b14 · outbound

This paper cites Introducing Claude Haiku 4.5.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing Claude Haiku 4.5

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.388224Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00fe7b66-8fb3-4a60-b459-90d41d47cdc4 · outbound

This paper cites Introducing Claude Opus 4.6.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing Claude Opus 4.6

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.568678Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc994809-00b1-4113-a0d8-727255b33736 · outbound

This paper cites Introducing Claude Sonnet 4.6.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing Claude Sonnet 4.6

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.702350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42d50484-fd84-4d72-bb1f-45c4d67beb0d · outbound

This paper cites Qwen3-VL Technical Report.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Qwen3-VL Technical Report

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.846358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a322ee7-bd9c-43d0-8fb1-c815e3775b67 · outbound

This paper cites Qwen2.5-VL Technical Report.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Qwen2.5-VL Technical Report

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:22.982366Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6704d8d7-3840-44fe-a0a4-fcc9e4296d58 · outbound

This paper cites OmnixR: Evaluating Omni-modality Language Models on Reasoning across Modalities.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs OmnixR: Evaluating Omni-modality Language Models on Reasoning across Modalities

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.079089Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9728877-6b05-4fac-836d-8c8630b34d32 · outbound

This paper cites Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.218254Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b5de65c-cc3c-4a3d-a736-808e8970bbc6 · outbound

This paper cites Reproducible scaling laws for contrastive language-image learning.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Reproducible scaling laws for contrastive language-image learning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.365501Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ab08e30-816b-402c-9b67-834b125bf141 · outbound

This paper cites Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.535097Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20e46a5d-9c0b-4505-9b39-08dc69a291c2 · outbound

This paper cites Mitigate the Gap: Investigating Approaches for Improving Cross-Modal Alignment in CLIP.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Mitigate the Gap: Investigating Approaches for Improving Cross-Modal Alignment in CLIP

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.673276Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ee249f0b-2bac-43a7-aee1-5d80eed4bd09 · outbound

This paper cites It's Not a Modality Gap: Characterizing and Addressing the Contrastive Gap.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs It's Not a Modality Gap: Characterizing and Addressing the Contrastive Gap

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.784825Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc9a9875-1f58-44a3-8bfb-79d9dd641b67 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:23.885090Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f539f254-2506-4a97-b330-42f890d0f58a · outbound

This paper cites Datacomp: In search of the next generation of multimodal datasets.Advances in Neural Information Processing Systems, 36:27092–27112, 2023.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Datacomp: In search of the next generation of multimodal datasets.Advances in Neural Information Processing Systems, 36:27092–27112, 2023

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.035488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 566e1361-173d-43c9-8ab6-142c4e7aaa54 · outbound

This paper cites Gemini API Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Gemini API Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.111180Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4f45055a-0d49-435c-a375-3294185d348b · outbound

This paper cites Measuring Massive Multitask Language Understanding.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Measuring Massive Multitask Language Understanding

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.238463Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da25ee71-5af0-4d47-affb-c6f345106c52 · outbound

This paper cites GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.375267Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation adbf1c56-c270-4099-918e-c7efecfbb718 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Gqa: A new dataset for real-world visual reasoning and compositional question answering

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.586305Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1de3104f-f5af-406c-816f-21b71a80e72d · outbound

This paper cites Scaling up visual and vision-language representation learning with noisy text supervision.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Scaling up visual and vision-language representation learning with noisy text supervision

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.718228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f529eec-e8cb-4847-a198-10bc23fba8ef · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs LLaVA-OneVision: Easy Visual Task Transfer

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.874739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8975576b-32f3-4353-9b1e-dc1c66504366 · outbound

This paper cites Mind the gap: Understanding the modality gap in multi-modal contrastive representation learning.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Mind the gap: Understanding the modality gap in multi-modal contrastive representation learning

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:24.983000Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2142a2af-4d36-44b3-9c7d-7975c404ab32 · outbound

This paper cites Improved baselines with visual instruction tuning.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Improved baselines with visual instruction tuning

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.149705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 012b3093-ea5f-4bdc-85bd-f04864a493b1 · outbound

This paper cites Mmbench: Is your multi-modal model an all-around player? InEuropean conference on computer vision, pages 216–233.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Mmbench: Is your multi-modal model an all-around player? InEuropean conference on computer vision, pages 216–233

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.276883Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86dc6205-2485-4ff5-8862-d7a54f39269d · outbound

This paper cites Scene text recognition using higher order language priors.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Scene text recognition using higher order language priors

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.361355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7ea08f0-a96b-48ba-a685-82f1f60e7849 · outbound

This paper cites GPT-4o System Card.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs GPT-4o System Card

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.450674Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4fc60180-ba70-4308-8b01-b080645a3c73 · outbound

This paper cites Introducing GPT-4.1 in the API.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing GPT-4.1 in the API

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.639465Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4f89df5b-97bc-4ea7-a589-c765dbdefef6 · outbound

This paper cites Introducing GPT-5.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing GPT-5

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.735867Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 436fbab9-e63f-4f85-8eb8-147172e5f104 · outbound

This paper cites Introducing GPT-5.1 for developers.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing GPT-5.1 for developers

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.804458Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ec1c4cd-3148-4c4e-a76a-c064e3633caa · outbound

This paper cites Introducing GPT-5.2.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Introducing GPT-5.2

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.899217Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0cfd5b92-4b19-4b07-9f29-0776e74f1302 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Learning transferable visual models from natural language supervision

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:25.999038Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00f0db15-6aa4-411f-a033-4e59e2061d5e · outbound

This paper cites Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.037233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 274f2c6e-a451-4eb8-a31d-569a5b5476c5 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.080641Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c2ba4d4-a67a-41a8-abd2-5f8ec0eebcfc · outbound

This paper cites Understanding the modality gap in clip.ICLR, Stockholm, Sweden, pages 8–10, 2023.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Understanding the modality gap in clip.ICLR, Stockholm, Sweden, pages 8–10, 2023

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.122685Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2486b27-fd6e-437c-aea4-920b8f8e2bec · outbound

This paper cites Towards vqa models that can read.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Towards vqa models that can read

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.166125Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation efc69535-bf0e-4bd0-8bd8-f5406d23872f · outbound

This paper cites EVA-CLIP: Improved Training Techniques for CLIP at Scale.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs EVA-CLIP: Improved Training Techniques for CLIP at Scale

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.204914Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 351a9023-0a61-4de3-8da0-72858a7b8fe0 · outbound

This paper cites SEAM: Semantically Equivalent Across Modalities Benchmark for Vision-Language Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs SEAM: Semantically Equivalent Across Modalities Benchmark for Vision-Language Models

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.251613Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 217acad4-03c2-4acb-bdf2-ffef09a28bf1 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Gemini: A Family of Highly Capable Multimodal Models

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.298996Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 795124b8-4d41-470b-a4a5-3e20985a1bf5 · outbound

This paper cites SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.353484Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c707c0d3-f389-4d8f-8197-e927dbff2e72 · outbound

This paper cites Same Content, Different Answers: Cross-Modal Inconsistency in MLLMs.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Same Content, Different Answers: Cross-Modal Inconsistency in MLLMs

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.402402Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1eabe3f4-4d68-4aa1-a539-42d62e00c877 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.464199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ebbec044-eab8-4e3c-8dd3-ceeac2ed0673 · outbound

This paper cites XModBench: Benchmarking Cross-Modal Capabilities and Consistency in Omni-Language Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs XModBench: Benchmarking Cross-Modal Capabilities and Consistency in Omni-Language Models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.517339Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ca04334-b4c5-4f82-901d-24312791a546 · outbound

This paper cites Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.568021Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b1609f3-bfac-4722-875a-be9f74c531c6 · outbound

This paper cites Explaining and mitigating the modality gap in contrastive multimodal learning.arXiv preprint arXiv:2412.07909, 2024.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Explaining and mitigating the modality gap in contrastive multimodal learning.arXiv preprint arXiv:2412.07909, 2024

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.635411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e4f6912-b653-4bc9-8eac-cd853d9e8242 · outbound

This paper cites Mmmu: A massive multi-discipline multimodal understanding and reasoning benchmark for expert agi.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Mmmu: A massive multi-discipline multimodal understanding and reasoning benchmark for expert agi

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.669492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62d4f79f-429e-4acf-9239-635abca262b9 · outbound

This paper cites MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.741382Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 45c5474f-a472-4e98-a4de-1637082407b5 · outbound

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

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Sigmoid loss for language image pre-training

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.794791Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74446a3f-8d0e-4e36-ba2a-d031d5df0af7 · outbound

This paper cites Lost in Translation: When GPT-4V(ision) Can't See Eye to Eye with Text. A Vision-Language-Consistency Analysis of VLLMs and Beyond.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Lost in Translation: When GPT-4V(ision) Can't See Eye to Eye with Text. A Vision-Language-Consistency Analysis of VLLMs and Beyond

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.849480Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a8925eb-0edb-4761-9cb5-c604d27d11c5 · outbound

This paper cites Cross-Modal Consistency in Multimodal Large Language Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Cross-Modal Consistency in Multimodal Large Language Models

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.897977Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8150157-dfc4-48d9-b641-ca7572dff2f1 · outbound

This paper cites InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.936954Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce086900-7c1e-4b2e-b0ed-6ecb8c0fa0cc · outbound

This paper cites - RETURN FALSE IF: The word is a verb but the image shows a noun (e.g., ’die’ as in death vs.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs - **RETURN FALSE IF**: The word is a verb but the image shows a noun (e.g., ’die’ as in death vs

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:26.982981Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 36b1e162-d94f-4555-9544-6780d6379617 · outbound

This paper cites human head.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs human head

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:27.028042Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation defab8ea-7503-4279-af6a-33b0d096f24f · outbound

This paper cites hot desert.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs hot desert

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:27.061895Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f74b7389-febf-4d15-9008-3199f1dcac5a · outbound

This paper cites an unresolved cited work.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Unresolved cited work

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:27.106744Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1462c91f-f99c-4c5d-aaf1-14bea1506142 · outbound

This paper cites The hot <image> (sand) scorched his feet.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs The hot <image> (sand) scorched his feet

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:27.183439Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2c58c819-1f7a-4a70-bb93-684284bb516c · outbound

This paper cites XModBench [41] further extends cross-modal consistency evaluation to tri-modal settings involving text, vision, and audio.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs XModBench [41] further extends cross-modal consistency evaluation to tri-modal settings involving text, vision, and audio

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-03T00:55:27.313147Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6958987-e9fb-4823-ad12-a23b65922f55 · outbound

This paper cites an unresolved cited work.

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs Unresolved cited work

Reference 2025

Resolution: parse uncertain; no resolver link, observed 2026-08-03T00:55:25.958444Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers