Paper Citation Record · LEDGER

Towards understanding how attention mechanism works in deep learning

As of 12 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 6 inbound Pith citation observations for arXiv:2412.18288.

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

pith.paper-citation-record.v1
2412.18288 v1

Coverage vector

measured 20 of 20 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T04:53:52.304925Z

measured 26 of 26 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-12T06:34:41.77262+00:00

measured 6 of 6 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T21:05:02.502481Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-01T21:56:15.441801Z

Reference resolution

20 of 20 outbound references displayed

verified exact1
verified fuzzy8
unresolved10
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 2f13138d-311a-40f9-8657-5c4d8e79a5a8 · outbound

This paper cites Proof: Denote the matrix of exp {fθ(xi, xj)} by W.

Towards understanding how attention mechanism works in deep learning Proof: Denote the matrix of exp {fθ(xi, xj)} by W

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.621405Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 138ac479-0c9f-4912-a910-d84cdb8ba9ba · outbound

This paper cites We conducted all experiments on a desktop computer with NVIDIA 2080Ti and 3.8 GHz AMD Ryzen 7 5800X 8-Core Process and 16 GB of memory and a computer with NVIDIA 3090Ti.

Towards understanding how attention mechanism works in deep learning We conducted all experiments on a desktop computer with NVIDIA 2080Ti and 3.8 GHz AMD Ryzen 7 5800X 8-Core Process and 16 GB of memory and a computer with NVIDIA 3090Ti

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.651971Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation c84a1717-e171-4b54-a7a6-394f2527b763 · outbound

This paper cites A mathematical perspective on Transformers.

Towards understanding how attention mechanism works in deep learning A mathematical perspective on Transformers

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.247956Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1aef1b6c-3581-4434-bbbe-2b2ca271c7e7 · outbound

This paper cites UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.

Towards understanding how attention mechanism works in deep learning UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.257258Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd099c74-af2d-4578-beda-60b558401f39 · outbound

This paper cites Graph Attention Networks.

Towards understanding how attention mechanism works in deep learning Graph Attention Networks

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.267529Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0e99909-dadb-4cf5-9784-4e12ffffaf4a · outbound

This paper cites URL http://dx.doi.org/10.18653/v1/2023.findings-acl.

Towards understanding how attention mechanism works in deep learning URL http://dx.doi.org/10.18653/v1/2023.findings-acl

Reference 15

Resolution: malformed identifier; no resolver link, observed 2026-08-11T04:53:52.281360Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a4d7c30-78d5-4f86-ac62-e1cd680c8a91 · outbound

This paper cites Experiments and results T oy dataset We evaluated the performance of metric-attention by comparing it with self-attention and L2 self-attention using the Moon dataset.

Towards understanding how attention mechanism works in deep learning Experiments and results T oy dataset We evaluated the performance of metric-attention by comparing it with self-attention and L2 self-attention using the Moon dataset

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.668163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 4508e9b1-f2a9-4128-b533-6fbaf5af89bd · outbound

This paper cites We adapted the implementation for the Multi30k dataset from https://github.com/hyunwoongko/transformer.

Towards understanding how attention mechanism works in deep learning We adapted the implementation for the Multi30k dataset from https://github.com/hyunwoongko/transformer

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.635416Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation bad485c5-fcff-4998-ae70-c902cac0c231 · outbound

This paper cites Manifold Fitting under Unbounded Noise.

Towards understanding how attention mechanism works in deep learning Manifold Fitting under Unbounded Noise

Reference 482

Resolution: verified exact; local_arxiv, observed 2026-08-11T04:53:52.373924Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 611a4699-9001-492a-b267-f481b1831bd0 · outbound

This paper cites Deep Residual Networks Learn the Geodesic Curve in the Wasserstein Space.

Towards understanding how attention mechanism works in deep learning Deep Residual Networks Learn the Geodesic Curve in the Wasserstein Space

Reference 1985

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.243199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ddbd2e32-7146-4b56-a5b2-1a9bf150283c · outbound

This paper cites doi: https://doi.

Towards understanding how attention mechanism works in deep learning doi: https://doi

Reference 1987

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.277386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c56f8573-1ad9-4b88-a1ef-60f37596e575 · outbound

This paper cites Score-Based Generative Modeling through Stochastic Differential Equations.

Towards understanding how attention mechanism works in deep learning Score-Based Generative Modeling through Stochastic Differential Equations

Reference 2006

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.262557Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a82944c5-dcb1-44ad-ae9b-4fc8c7682e65 · outbound

This paper cites A Mathematical Theory of Attention.

Towards understanding how attention mechanism works in deep learning A Mathematical Theory of Attention

Reference 2008

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.272809Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e87404cd-b3f6-4a3a-bfdf-a0ad21ef621e · outbound

This paper cites Lan- guage models are few-shot learners.

Towards understanding how attention mechanism works in deep learning Lan- guage models are few-shot learners

Reference 2010

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.720224Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation a1cc6f57-d9fd-4fef-9c04-c32af807122d · outbound

This paper cites BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.

Towards understanding how attention mechanism works in deep learning BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 2012

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.221496Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b06985d4-25c0-4357-8667-253b52eac369 · outbound

This paper cites Scale-invariant heat kernel signatures for non- rigid shape recognition.

Towards understanding how attention mechanism works in deep learning Scale-invariant heat kernel signatures for non- rigid shape recognition

Reference 2014

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.733661Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 8d2806b5-461b-4c3a-8637-bdc8401350fd · outbound

This paper cites Multi30K: Multilingual English-German Image Descriptions.

Towards understanding how attention mechanism works in deep learning Multi30K: Multilingual English-German Image Descriptions

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.237619Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cefb950e-77b6-4e3f-ad29-dbdfc3a6a3f0 · outbound

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

Towards understanding how attention mechanism works in deep learning An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-11T04:53:52.232364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e47f16bc-e2a8-4ea4-bb8e-8ec16c936997 · outbound

This paper cites Speech-transformer: A no-recurrence sequence- to-sequence model for speech recognition.

Towards understanding how attention mechanism works in deep learning Speech-transformer: A no-recurrence sequence- to-sequence model for speech recognition

Reference 2022

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.703501Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 114cfe83-b128-435b-9dc4-0ad77db2f974 · outbound

This paper cites Shape retrieval contest 2007: Wa- tertight models track.

Towards understanding how attention mechanism works in deep learning Shape retrieval contest 2007: Wa- tertight models track

Reference 2023

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T04:53:52.684496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Pith citing papers

Observation fe4b9dc8-3e46-43ed-9d3a-6f8ce3ca128e · inbound

Physics- and geometry-aware spatio-spectral graph neural operator for time-independent and time-dependent PDEs cites this paper.

Physics- and geometry-aware spatio-spectral graph neural operator for time-independent and time-dependent PDEs Towards understanding how attention mechanism works in deep learning

Reference 2012

Resolution: unresolved; no resolver link, observed 2026-08-05T20:59:36.897052Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02f2568a-cf97-47ad-8fe9-c5fa3b1a24af · inbound

Attention's forward pass and Frank-Wolfe cites this paper.

Attention's forward pass and Frank-Wolfe Towards understanding how attention mechanism works in deep learning

Reference 2012

Resolution: unresolved; no resolver link, observed 2026-08-05T21:05:02.502481Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5745ca25-4c14-4d87-9b02-61d40e725aa0 · inbound

Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation cites this paper.

Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation Towards understanding how attention mechanism works in deep learning

Reference 74

Resolution: verified exact; arxiv_id, observed 2026-05-18T19:01:46.313882Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation bf9ec9b5-68dd-413d-ac48-f943d92837ce · inbound

Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation cites this paper.

Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation Towards understanding how attention mechanism works in deep learning

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-05T10:25:17.199621Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1740aba-fbd5-4a2a-bea8-58baccbf15cd · inbound

MomentKV: Closing the Directional Gap in KV Cache Eviction for Long-Context Inference cites this paper.

MomentKV: Closing the Directional Gap in KV Cache Eviction for Long-Context Inference Towards understanding how attention mechanism works in deep learning

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-07-01T21:56:15.443916Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 4368ff3f-46d2-4554-b36c-331c141bbd05 · inbound

From Self-Attention to Connection Laplacian: A Unified Operator View of Transformers cites this paper.

From Self-Attention to Connection Laplacian: A Unified Operator View of Transformers Towards understanding how attention mechanism works in deep learning

Reference 20

Resolution: unresolved; no resolver link, observed 2026-07-14T10:03:27.747512Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.