Paper Citation Record · LEDGER

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention

As of 22 August 2026, this Paper Citation Record lists 23 of 23 outbound references and 1 inbound Pith citation observation for arXiv:2507.09394.

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

pith.paper-citation-record.v1
2507.09394 v1

Coverage vector

measured 23 of 23 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T18:07:38.566888Z

measured 24 of 24 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-22T06:32:14.747728+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-11T00:52:08.399401Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T05:05:58.687554Z

Reference resolution

23 of 23 outbound references displayed

verified exact0
verified fuzzy16
unresolved7
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation ee683523-97b9-4efe-a703-6c44ed8cdb82 · outbound

This paper cites A random matrix perspective on mix- tures of nonlinearities in high dimensions.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention A random matrix perspective on mix- tures of nonlinearities in high dimensions

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:41.900833Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8cf650c4-73bd-4dc8-86ba-fab5670b39a9 · outbound

This paper cites Self-attention networks localize when QK- eigenspectrum concentrates.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Self-attention networks localize when QK- eigenspectrum concentrates

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:41.736678Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3ee33393-c328-446e-824b-1de74fadeb71 · outbound

This paper cites Random matrix theory improved fr ´echet mean of symmetric positive definite matrices.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Random matrix theory improved fr ´echet mean of symmetric positive definite matrices

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:41.568076Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 349b43ec-d02e-4ff6-83fb-abf279ddd3c0 · outbound

This paper cites A random matrix ap- proach to echo-state neural networks.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention A random matrix ap- proach to echo-state neural networks

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:41.319175Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 81c403de-6cd6-47f2-a51c-71fb29dd1ce1 · outbound

This paper cites A random matrix theory perspective on the spectrum of learned features and asymptotic generalization capabilities.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention A random matrix theory perspective on the spectrum of learned features and asymptotic generalization capabilities

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:41.118098Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation a11f441f-56f7-4e0d-9000-0ac68f94eb44 · outbound

This paper cites Random matrix analysis to balance between supervised and unsupervised learning under the low density separation assumption.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Random matrix analysis to balance between supervised and unsupervised learning under the low density separation assumption

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.943455Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 98dffd79-1563-419d-923b-573a5551b30b · outbound

This paper cites Maximizing the potential of synthetic data: Insights from ran- dom matrix theory.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Maximizing the potential of synthetic data: Insights from ran- dom matrix theory

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.719825Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 45bcfe2a-a32e-45dd-8d56-451575538272 · outbound

This paper cites Analysing multi-task regression via random matrix theory with application to time series forecasting.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Analysing multi-task regression via random matrix theory with application to time series forecasting

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.541879Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 25ea31cf-a862-4943-97aa-7e3c08b5dcae · outbound

This paper cites The Underlying Scaling Laws and Universal Statistical Structure of Complex Datasets.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention The Underlying Scaling Laws and Universal Statistical Structure of Complex Datasets

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:37.592873Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8971b61d-d2b0-4d5a-b90d-7d9437f98b6b · outbound

This paper cites Mix-LN: Unleashing the power of deeper layers by combining pre-LN and post-LN.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Mix-LN: Unleashing the power of deeper layers by combining pre-LN and post-LN

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.349433Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation c6ac8cd5-f5f1-4af1-8e41-5c10500bd25b · outbound

This paper cites The dynamics of learning: A random matrix approach.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention The dynamics of learning: A random matrix approach

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.148126Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 2b056b2e-0d9d-42bc-877e-fa12dc11cac8 · outbound

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

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:37.781457Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8c6a1f5a-6fe6-46b8-be9a-d5c50b87b7b5 · outbound

This paper cites DeepSeek-V3 Technical Report.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention DeepSeek-V3 Technical Report

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:37.848948Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 230f14ec-31ac-4344-95e8-7acaf51b5acc · outbound

This paper cites Distribution of eigenvalues for some sets of random matrices.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Distribution of eigenvalues for some sets of random matrices

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:40.021515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 43395909-49b0-4b56-84a4-8915f5ab0f54 · outbound

This paper cites Implicit self-regularization in deep neural net- works: Evidence from random matrix theory and implications for learning.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Implicit self-regularization in deep neural net- works: Evidence from random matrix theory and implications for learning

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:39.837235Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation db4c0ef7-315a-436c-89d6-569d992d4d08 · outbound

This paper cites TransMLA: Multi-Head Latent Attention Is All You Need.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention TransMLA: Multi-Head Latent Attention Is All You Need

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:38.019228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cb10a6e6-f7da-4471-82ed-b3178aad6735 · outbound

This paper cites Geometry of neural network loss surfaces via random matrix theory.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Geometry of neural network loss surfaces via random matrix theory

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:39.600408Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f8ca5be1-61db-472c-a785-0e266df756c9 · outbound

This paper cites Nonlinear random matrix theory for deep learning.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Nonlinear random matrix theory for deep learning

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:38.183231Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2aa809c4-6732-48d9-af69-59694030d5fe · outbound

This paper cites Locating information in large language models via random matrix theory.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Locating information in large language models via random matrix theory

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:38.258595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f33cf979-feb5-402e-82c3-4c4caf4073d4 · outbound

This paper cites Random matrix theory analysis of neural network weight matrices.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Random matrix theory analysis of neural network weight matrices

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:39.412438Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 36013000-9d1b-44b9-aaea-1875b88768bc · outbound

This paper cites Random ma- trix improved covariance estimation for a large class of metrics.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Random ma- trix improved covariance estimation for a large class of metrics

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:39.199843Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5cd7fe0c-24b6-48c5-b213-3872f97779b1 · outbound

This paper cites More than a toy: Random matrix models pre- dict how real-world neural representations generalize.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention More than a toy: Random matrix models pre- dict how real-world neural representations generalize

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:07:39.022322Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4d42d40f-dc65-4ff9-b0d8-1b5136db0d19 · outbound

This paper cites Insights into deepseek-v3: Scaling challenges and reflections on hardware for ai architectures.

A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention Insights into deepseek-v3: Scaling challenges and reflections on hardware for ai architectures

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-06T18:07:38.566888Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 51d386c9-4ea1-4944-a112-cfc16267f99a · inbound

How Does Attention Help? Insights from Random Matrices on Signal Recovery from Sequence Models cites this paper.

How Does Attention Help? Insights from Random Matrices on Signal Recovery from Sequence Models A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-11T05:05:58.694020Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.