Paper Citation Record · LEDGER

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

As of 9 August 2026, this Paper Citation Record lists 64 of 64 outbound references and 9 inbound Pith citation observations for arXiv:2502.12170.

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

pith.paper-citation-record.v1
2502.12170 v2

Coverage vector

measured 64 of 64 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T22:34:08.193663Z

measured 73 of 73 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 9 of 9 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-06T00:37:19.285243Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-02T21:57:25.819786Z

Reference resolution

64 of 64 outbound references displayed

verified exact2
verified fuzzy23
unresolved39
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 78de3437-f83e-4a79-854e-3c548bfa6b33 · outbound

This paper cites write newline.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.945288Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5fbc684-6e16-46ca-891f-44d77627d184 · outbound

This paper cites G., Bradley, H., O’Brien, K., Hallahan, E., Khan, M.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections G., Bradley, H., O’Brien, K., Hallahan, E., Khan, M

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.115672Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 0a07bd64-94da-4d53-9c52-87f15624e263 · outbound

This paper cites Piqa: Reasoning about physical commonsense in natural language.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Piqa: Reasoning about physical commonsense in natural language

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.957195Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9bbe0a22-fc85-4ac7-81a7-82c15fcbdcf7 · outbound

This paper cites Reducing Transformer Key-Value Cache Size with Cross-Layer Attention.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Reducing Transformer Key-Value Cache Size with Cross-Layer Attention

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.961504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation adc8c649-c9b5-4787-99cf-e498c7f9f4db · outbound

This paper cites D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.966880Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb6f8e41-dfdf-4204-b314-fed67fe98ee5 · outbound

This paper cites BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.971082Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6a1af4f2-b42d-4e90-829c-8ebb0d862364 · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.975673Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 90e9047c-d70e-468b-99ff-0afd5130bd58 · outbound

This paper cites Towards automated circuit discovery for mechanistic interpretability.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Towards automated circuit discovery for mechanistic interpretability

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.092559Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation e793411a-af5e-4a98-b510-8ac86833f645 · outbound

This paper cites DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.985133Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 281fcbeb-e5f9-4788-aeed-95e75f583154 · outbound

This paper cites and Gu, A.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections and Gu, A

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.081902Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 1fecb28e-c183-47c0-8bf3-6aeebe58ca2e · outbound

This paper cites An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:07.993015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 559754bf-8d42-4c2e-ba1d-cd3db4dc1b31 · outbound

This paper cites A mathematical framework for transformer circuits.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections A mathematical framework for transformer circuits

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.071092Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5e110141-4496-4db4-b439-8e161a8d815d · outbound

This paper cites Depth-Wise Attention (DWAtt): A Layer Fusion Method for Data-Efficient Classification.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Depth-Wise Attention (DWAtt): A Layer Fusion Method for Data-Efficient Classification

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-08-07T22:34:08.604471Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 50861494-8d99-4dd8-88ad-9561955b5398 · outbound

This paper cites Cross-Layer Retrospective Retrieving via Layer Attention.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Cross-Layer Retrospective Retrieving via Layer Attention

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.005171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 276a375c-5f30-4ebd-aa38-097600dcb5fe · outbound

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

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.060104Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f459478a-7ab8-4093-bc94-b63a81853df2 · outbound

This paper cites The Pile: An 800GB Dataset of Diverse Text for Language Modeling.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The Pile: An 800GB Dataset of Diverse Text for Language Modeling

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.014923Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12848330-225c-4ad9-9865-585b4bb6673a · outbound

This paper cites A framework for few-shot language model evaluation, 12 2023.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections A framework for few-shot language model evaluation, 12 2023

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.018872Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aed77fdf-3f4e-48ae-b846-e454c23d7916 · outbound

This paper cites The Unreasonable Ineffectiveness of the Deeper Layers.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The Unreasonable Ineffectiveness of the Deeper Layers

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.022953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0049f25d-2b5d-4371-a861-df433f518b4b · outbound

This paper cites and Dao, T.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections and Dao, T

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.048522Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation e7e781c3-542c-47c1-be1a-24ebc64c05d9 · outbound

This paper cites Have faith in faithfulness: Going beyond circuit overlap when finding model mechanisms.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Have faith in faithfulness: Going beyond circuit overlap when finding model mechanisms

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.036957Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation bc7b820d-e495-42ce-9252-8031d17125bf · outbound

This paper cites Deep residual learning for image recognition.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Deep residual learning for image recognition

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.025320Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 0ee51f95-b561-4f3a-914e-165c4e4bd41b · outbound

This paper cites A., Welbl, J., Clark, A., et al.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections A., Welbl, J., Clark, A., et al

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:09.014215Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation be8b1684-8a4b-4936-b619-2da0b7d502af · outbound

This paper cites an unresolved cited work.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-08-07T22:34:09.003950Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 93f41a94-7d94-4a30-a59c-77eaa2118d38 · outbound

This paper cites A., Casper, J., Lym, S., McAfee, L., Andersch, M., Shoeybi, M., and Catanzaro, B.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections A., Casper, J., Lym, S., McAfee, L., Andersch, M., Shoeybi, M., and Catanzaro, B

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.045980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 16b1a208-58e2-4fe2-9282-66839e427942 · outbound

This paper cites RACE: Large-scale ReAding Comprehension Dataset From Examinations.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections RACE: Large-scale ReAding Comprehension Dataset From Examinations

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.049712Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4580450f-51c6-4231-87ed-b6b090347ca0 · outbound

This paper cites Selective Attention Improves Transformer.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Selective Attention Improves Transformer

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.054013Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85c2f796-d36f-44bb-8a1a-21c61db8f5e2 · outbound

This paper cites DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.058138Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4b6e1c56-cd98-4ff7-a840-c313d7510089 · outbound

This paper cites DeepSeek-V3 Technical Report.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections DeepSeek-V3 Technical Report

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.062063Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2372c58-8453-4e7c-a9e4-b0c4e62e3e6a · outbound

This paper cites LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical Reasoning.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical Reasoning

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.066168Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 64c6889a-87fb-4cb0-9241-82ed38791d99 · outbound

This paper cites Understanding the difficulty of training transformers.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Understanding the difficulty of training transformers

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.987678Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation ea988790-656e-497c-ae5b-91ac55725b5a · outbound

This paper cites The Flan Collection: Designing Data and Methods for Effective Instruction Tuning.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The Flan Collection: Designing Data and Methods for Effective Instruction Tuning

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.076522Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 79f4fbe5-aa47-44a2-aef9-ade142bc0514 · outbound

This paper cites an unresolved cited work.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Unresolved cited work

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.080327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f7b3735-c0d4-4264-a896-f3ac38ce0eeb · outbound

This paper cites Talking Heads: Understanding Inter-layer Communication in Transformer Language Models.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Talking Heads: Understanding Inter-layer Communication in Transformer Language Models

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-08-07T22:34:08.355036Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f5b338dd-a66a-4811-b9f9-1b63c717b965 · outbound

This paper cites Olmoe: Open mixture-of-experts language models.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Olmoe: Open mixture-of-experts language models

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.969449Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 67e92589-2861-4072-b0ad-e19826aea01f · outbound

This paper cites Benchmarking and understanding compositional relational reasoning of llms.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Benchmarking and understanding compositional relational reasoning of llms

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.957373Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation cb86b5e1-0fea-400d-8745-1b6535edfb09 · outbound

This paper cites Denseformer: Enhancing information flow in transformers via depth weighted averaging.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Denseformer: Enhancing information flow in transformers via depth weighted averaging

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.945730Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 1a80463c-eb8f-41b7-8c6e-129879173865 · outbound

This paper cites The LAMBADA dataset: Word prediction requiring a broad discourse context.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The LAMBADA dataset: Word prediction requiring a broad discourse context

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.098145Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da1cca8d-f661-4991-9c81-ed796a1cb06d · outbound

This paper cites Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.101679Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de384c6d-2abf-4988-8da7-ea42cdb1a163 · outbound

This paper cites The Impact of Depth on Compositional Generalization in Transformer Language Models.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The Impact of Depth on Compositional Generalization in Transformer Language Models

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.105371Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 71a98e51-772d-4ac4-adc0-ddae9438ed28 · outbound

This paper cites L., Bhagavatula, C., and Choi, Y.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections L., Bhagavatula, C., and Choi, Y

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.108857Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fd6a5cd4-0d90-46df-9405-3d885690f51a · outbound

This paper cites GLU Variants Improve Transformer.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections GLU Variants Improve Transformer

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.112283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df5a1019-99f0-460e-837b-b49d6ada3470 · outbound

This paper cites K., Greff, K., and Schmidhuber, J.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections K., Greff, K., and Schmidhuber, J

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.927394Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 08d30a16-5875-475c-82f8-a6206fcf8ef6 · outbound

This paper cites Roformer: Enhanced transformer with rotary position embedding.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Roformer: Enhanced transformer with rotary position embedding

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.119680Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98fb475d-37a4-40ac-b6b0-cb78bc989ab2 · outbound

This paper cites M., Qin, Z., Bahri, D., Juan, D.-C., and Metzler, D.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections M., Qin, Z., Bahri, D., Juan, D.-C., and Metzler, D

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.908195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation ceb5ec8e-0c97-4438-a924-013435778864 · outbound

This paper cites Scale Efficiently: Insights from Pre-training and Fine-tuning Transformers.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Scale Efficiently: Insights from Pre-training and Fine-tuning Transformers

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.126213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 69d94b94-7826-4bf1-a151-f6f52c5565e8 · outbound

This paper cites Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.129759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d98f208-7f51-4f4c-8b82-6d81d1751385 · outbound

This paper cites Gemma 2: Improving Open Language Models at a Practical Size.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Gemma 2: Improving Open Language Models at a Practical Size

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.133254Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b3c6c22-a86f-4741-85d5-67ad3627c792 · outbound

This paper cites The llama 4 herd: The beginning of a new era of natively multimodal ai innovation.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections The llama 4 herd: The beginning of a new era of natively multimodal ai innovation

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.897753Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation d1c6490c-c50c-4449-bd60-45278570512c · outbound

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

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections LLaMA: Open and Efficient Foundation Language Models

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.139802Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07545191-d33a-4dfb-8d27-5028d9dd6aed · outbound

This paper cites N., Kaiser, ., and Polosukhin, I.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections N., Kaiser, ., and Polosukhin, I

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.144041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3590de0c-30e4-41bd-b9fc-6feaf80abdd6 · outbound

This paper cites Analyzing the Structure of Attention in a Transformer Language Model.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Analyzing the Structure of Attention in a Transformer Language Model

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.147780Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2dca6d98-564a-4188-a711-4b1f8bc7df8b · outbound

This paper cites Interpretability in the wild: a circuit for indirect object identification in gpt-2 small.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Interpretability in the wild: a circuit for indirect object identification in gpt-2 small

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.756353Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 782363fe-2d79-4f88-9471-7e0400c1fe5b · outbound

This paper cites Strengthening Layer Interaction via Dynamic Layer Attention.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Strengthening Layer Interaction via Dynamic Layer Attention

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.154891Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6da75b33-2b62-46ca-a5ee-91d2c3d78484 · outbound

This paper cites F., and Chao, L.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections F., and Chao, L

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.745509Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 820fd855-2e90-4726-a66b-79da05a91bc8 · outbound

This paper cites Emergent Abilities of Large Language Models.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Emergent Abilities of Large Language Models

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.161634Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0fdb3c81-5243-4113-afce-bd0beb9d1afb · outbound

This paper cites Crowdsourcing Multiple Choice Science Questions.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Crowdsourcing Multiple Choice Science Questions

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.165072Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20c44499-9d8b-4888-940a-cecb5c7dc5b7 · outbound

This paper cites an unresolved cited work.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Unresolved cited work

Reference 57

Resolution: unresolved; raw_fallback, observed 2026-08-07T22:34:08.734317Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation da90a5c5-9ae8-43e4-9f86-7ad518e8b4c6 · outbound

This paper cites Improving transformers with dynamically composable multi-head attention.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Improving transformers with dynamically composable multi-head attention

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.722852Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 0b3f34d1-5665-49be-9bdf-343f0f2cbed7 · outbound

This paper cites Efficient streaming language models with attention sinks.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Efficient streaming language models with attention sinks

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.711762Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5f0a707f-3cf4-4954-89e9-657bb99a6bf1 · outbound

This paper cites Qwen3 Technical Report.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Qwen3 Technical Report

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.179053Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a50b0f43-d8c6-4160-9af6-14990bae22da · outbound

This paper cites Parallelizing linear transformers with the delta rule over sequence length.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Parallelizing linear transformers with the delta rule over sequence length

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.699670Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 08af5090-fac2-4b47-b718-992219d0effe · outbound

This paper cites Differential transformer.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Differential transformer

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.687836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 4564dbfb-7ec1-478d-b831-b19f351df87e · outbound

This paper cites HellaSwag: Can a Machine Really Finish Your Sentence?.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections HellaSwag: Can a Machine Really Finish Your Sentence?

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-07T22:34:08.189878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c492a069-11a9-4ba0-b8e3-9b3dc6786352 · outbound

This paper cites Hyper-connections.

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections Hyper-connections

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T22:34:08.674949Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Pith citing papers

Observation 2d3552d9-68d0-49d1-9622-d24430cf7d15 · inbound

mHC: Manifold-Constrained Hyper-Connections cites this paper.

mHC: Manifold-Constrained Hyper-Connections MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-15T12:31:48.052805Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation e51a55f0-113c-402c-b5a4-fc584766dc86 · inbound

KromHC: Manifold-Constrained Hyper-Connections with Kronecker-Product Residual Matrices cites this paper.

KromHC: Manifold-Constrained Hyper-Connections with Kronecker-Product Residual Matrices MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T06:57:48.680354Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 405e42fb-79ed-4aa9-925b-71c41d795c7a · inbound

Hyperloop Transformers cites this paper.

Hyperloop Transformers MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 26

Resolution: verified exact; arxiv_id, observed 2026-05-11T14:16:04.479032Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5834727e-3a57-4431-b054-8749a49b4191 · inbound

Transformers with Selective Access to Early Representations cites this paper.

Transformers with Selective Access to Early Representations MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-05-11T23:46:54.688717Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation ed8c9f0b-f56a-4604-be68-c444d2bd9148 · inbound

Transformers with Selective Access to Early Representations cites this paper.

Transformers with Selective Access to Early Representations MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-05-09T06:45:41.024855Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation cbe93a6b-f7b7-4ff5-adfb-bd48f8a72311 · inbound

Depth-Attention: Cross-Layer Value Mixing for Language Models cites this paper.

Depth-Attention: Cross-Layer Value Mixing for Language Models MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 13

Resolution: metadata mismatch; arxiv_id, observed 2026-07-02T08:36:48.483057Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 6c545116-5285-42e3-ba19-d015910dc2f9 · inbound

DeRes: Decoupling Residual Stability and Adaptivity for Scalable CTR Prediction cites this paper.

DeRes: Decoupling Residual Stability and Adaptivity for Scalable CTR Prediction MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 38

Resolution: verified exact; arxiv_id, observed 2026-07-02T21:57:25.821381Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f4142e81-0cb3-4a97-ba25-defe4e4fda75 · inbound

Enhancing Layer Interaction Using Key-Correlated Layer Attention cites this paper.

Enhancing Layer Interaction Using Key-Correlated Layer Attention MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 26

Resolution: metadata mismatch; arxiv_id, observed 2026-07-01T15:45:48.451446Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 1edb3645-7ef2-4d43-938c-ea74b5a4f564 · inbound

Role-Decoupled Attention Residuals: Separating Matching and Content Retrieval Across Depth cites this paper.

Role-Decoupled Attention Residuals: Separating Matching and Content Retrieval Across Depth MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-06T00:37:19.285243Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.