Paper Citation Record · LEDGER

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs

As of 15 August 2026, this Paper Citation Record lists 21 of 21 outbound references and 1 inbound Pith citation observation for arXiv:2605.22297.

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

pith.paper-citation-record.v1
2605.22297 v3

Coverage vector

measured 21 of 21 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-11T11:50:26.030339Z

measured 22 of 22 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-15T06:32:42.880941+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-02T01:58:56.074206Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

21 of 21 outbound references displayed

verified exact15
verified fuzzy1
unresolved0
parse uncertain0
malformed identifier1
metadata mismatch4

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 819e1380-7647-451f-85fc-da480930b13f · outbound

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

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.584086Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation cbe0943e-ee89-4248-9b27-fbfc0a7c29cd · outbound

This paper cites Don’t be lazy: Completep enables compute-efficient deep transformers.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Don’t be lazy: Completep enables compute-efficient deep transformers

Reference 2

Resolution: metadata mismatch; arxiv_id, observed 2026-06-30T17:34:57.587016Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation bc462715-4679-4150-bef0-3cee690c782f · outbound

This paper cites Optimal Embedding Learning Rate in LLMs: The Effect of Vocabulary Size.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Optimal Embedding Learning Rate in LLMs: The Effect of Vocabulary Size

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.595450Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation f9de646d-499a-4748-9436-1d405e734dfd · outbound

This paper cites Alphadecay: Module-wise weight decay for heavy-tailed balancing in llms.arXiv preprint arXiv:2506.14562.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Alphadecay: Module-wise weight decay for heavy-tailed balancing in llms.arXiv preprint arXiv:2506.14562

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.601205Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 67fc3bb7-d4f8-4a94-9474-7e88faef9e28 · outbound

This paper cites Training Compute-Optimal Large Language Models.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Training Compute-Optimal Large Language Models

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.633293Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation fe5e839b-61a3-46ef-a501-15c2a0fe2df3 · outbound

This paper cites LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.589775Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 400b49a7-d530-4f35-8c50-7b2f34679cef · outbound

This paper cites Muon is Scalable for LLM Training.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Muon is Scalable for LLM Training

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.604761Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 15ab6c0a-27ba-4ae0-8781-557c59d4bbb8 · outbound

This paper cites Model Balancing Helps Low-data Training and Fine-tuning.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Model Balancing Helps Low-data Training and Fine-tuning

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.627601Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 335f0174-667c-4858-9817-ade1ff1f7da0 · outbound

This paper cites Decoupled Weight Decay Regularization.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Decoupled Weight Decay Regularization

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.624687Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 2c26e020-6f35-494f-bbb9-bcd4e03b191f · outbound

This paper cites Traditional and Heavy-Tailed Self Regularization in Neural Network Models.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Traditional and Heavy-Tailed Self Regularization in Neural Network Models

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.621695Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation cec8cb3f-1977-48a2-b2df-caee9a25e30b · outbound

This paper cites Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.630682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 1be24f73-fb06-45b3-82d2-97602b6189a3 · outbound

This paper cites The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale

Reference 12

Resolution: metadata mismatch; local_arxiv, observed 2026-06-30T17:34:57.613690Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 5eb45495-a603-400a-bc8a-6c3873053e2a · outbound

This paper cites SocialIQA: Commonsense Reasoning about Social Interactions.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs SocialIQA: Commonsense Reasoning about Social Interactions

Reference 13

Resolution: metadata mismatch; local_arxiv, observed 2026-06-30T17:34:56.910283Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 8b49eecf-6cbd-4f43-9330-9c01c3b7f723 · outbound

This paper cites The Sharpness Disparity Principle in Transformers for Accelerating Language Model Pre-Training.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs The Sharpness Disparity Principle in Transformers for Accelerating Language Model Pre-Training

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.616526Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 1eb83793-db37-4efa-95da-fff07bd1647f · outbound

This paper cites Feature Learning in Infinite-Width Neural Networks.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Feature Learning in Infinite-Width Neural Networks

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-06-30T17:34:57.603774Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation e9a303f1-aa0d-4451-af4e-933cde503842 · outbound

This paper cites Large Batch Training of Convolutional Networks.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Large Batch Training of Convolutional Networks

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.606238Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation e173ea4f-b4a6-42d8-8117-72b6db0c3b36 · outbound

This paper cites Large Batch Optimization for Deep Learning: Training BERT in 76 minutes.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Large Batch Optimization for Deep Learning: Training BERT in 76 minutes

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.608787Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 1112d5bd-7908-4d8c-b9f0-4a1d17511620 · outbound

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

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs HellaSwag: Can a Machine Really Finish Your Sentence?

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.611239Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 7e02c1ac-30c9-489b-a4c3-8e5c92189eac · outbound

This paper cites Adam-mini: Use Fewer Learning Rates To Gain More.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Adam-mini: Use Fewer Learning Rates To Gain More

Reference 19

Resolution: metadata mismatch; arxiv_id, observed 2026-06-30T17:34:57.598365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 7f1bf2a3-b974-4d96-a975-cf417390f4bb · outbound

This paper cites Details of Experiments This section provides detailed configurations for both pre- training and finetuning experiments.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Details of Experiments This section provides detailed configurations for both pre- training and finetuning experiments

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-07-08T08:54:50.066199Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 02c5b434-0d25-44cd-b0b6-d74a456cb6bb · outbound

This paper cites Among these, Linear achieves the best results across all LR settings, showing a notable advantage over other methods.

One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs Among these, Linear achieves the best results across all LR settings, showing a notable advantage over other methods

Reference 21

Resolution: malformed identifier; arxiv_id, observed 2026-06-30T17:34:57.619301Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Pith citing papers

Observation 10f40f6a-933f-426d-ad80-6a7aac84ea9b · inbound

Muse: Representation Geometry of Muon Beyond Normalized Momentum cites this paper.

Muse: Representation Geometry of Muon Beyond Normalized Momentum One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T01:58:56.074206Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.