Paper Citation Record · LEDGER

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers

As of 10 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 2 inbound Pith citation observations for arXiv:2601.22580.

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

pith.paper-citation-record.v1
2601.22580 v2

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-03T06:37:55.054757Z

measured 26 of 26 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-02T03:08:35.290649Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

24 of 24 outbound references displayed

verified exact0
verified fuzzy0
unresolved23
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 293ba4ee-e339-4786-80db-b2dbe1f379f1 · outbound

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

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:51.875002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d237581-7563-45b5-bbde-8f33e8f20485 · outbound

This paper cites an unresolved cited work.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Unresolved cited work

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.174764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd10dc45-e6d9-413b-b4c2-10712ad8afb8 · outbound

This paper cites This isolates the impact of the architecture from the initialization scheme.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers This isolates the impact of the architecture from the initialization scheme

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.584809Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7fcc269-84a2-4d5e-a514-8f5e1515621f · outbound

This paper cites Li, P., Yin, L., and Liu, S.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Li, P., Yin, L., and Liu, S

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.874760Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc0dc7c0-85ff-46ac-a3ae-c82289339053 · outbound

This paper cites DeepSeek-V3 Technical Report.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers DeepSeek-V3 Technical Report

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.044794Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 092d188c-22f4-48ec-aecf-9283e1ad4ed6 · outbound

This paper cites Spike No More: Stabilizing the Pre-training of Large Language Models.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Spike No More: Stabilizing the Pre-training of Large Language Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.414766Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58e2e5ac-c5cc-49b1-b893-abb46450f71b · outbound

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

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Gemma 2: Improving Open Language Models at a Practical Size

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.551888Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0900031-97ac-4c3a-a19a-e25507a68749 · outbound

This paper cites L., Li, B., Lei, B., Wang, B., Rong, B., Wang, C., Zhang, C., Gao, C., Zhang, C., Sun, C., et al.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers L., Li, B., Lei, B., Wang, B., Rong, B., Wang, C., Zhang, C., Gao, C., Zhang, C., Sun, C., et al

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.644874Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6ea308e-b69c-4a77-896f-cb6f09727799 · outbound

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

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers LLaMA: Open and Efficient Foundation Language Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.764764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3238ff96-6bf5-48e8-8794-a9cf4ac718ef · outbound

This paper cites Qwen3 Technical Report.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Qwen3 Technical Report

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.934754Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6dce9cb-04ad-4ae2-85e6-515d2827fca3 · outbound

This paper cites Improving deep transformer with depth-scaled initialization and merged attention.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Improving deep transformer with depth-scaled initialization and merged attention

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.044772Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7399ff37-2bdb-439e-919d-6b77bb2df8cb · outbound

This paper cites 10 SpanNorm: Reconciling Training Stability and Performance in Deep Transformers A.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers 10 SpanNorm: Reconciling Training Stability and Performance in Deep Transformers A

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.204766Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c50c541-1c22-4056-8cee-43e8bfb14b7e · outbound

This paper cites an unresolved cited work.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Unresolved cited work

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.376442Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 302ab2e2-82a8-401b-87c3-80f181cbb33a · outbound

This paper cites In contrast, SpanNorm demonstrates superior representational capability.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers In contrast, SpanNorm demonstrates superior representational capability

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.684811Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfe41fac-244a-41c7-ae4f-97059b847ab6 · outbound

This paper cites All models are trained on the same subset of the SlimPajama dataset (from 30B to 200B) with the Mistral tokenizer Jiang et al.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers All models are trained on the same subset of the SlimPajama dataset (from 30B to 200B) with the Mistral tokenizer Jiang et al

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.934759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db33dd57-8bab-4de8-8000-12cb264d854e · outbound

This paper cites an unresolved cited work.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Unresolved cited work

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:55.054757Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 103a586a-32d8-4e97-b0bf-b6b0082c4c3c · outbound

This paper cites It employs an MLA architecture, activating 6 out of 64 experts.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers It employs an MLA architecture, activating 6 out of 64 experts

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:54.475088Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8aeaa3c-e490-4587-a8e8-cf716e482242 · outbound

This paper cites SpanNorm shows a strict monotonic decrease in training loss, confirming robust stability and the effective avoidance of depth degradation at extreme scales.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers SpanNorm shows a strict monotonic decrease in training loss, confirming robust stability and the effective avoidance of depth degradation at extreme scales

Reference 384

Resolution: malformed identifier; no resolver link, observed 2026-08-03T06:37:54.805213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 35c1985e-c86e-49eb-b860-1e88f053b8eb · outbound

This paper cites Shallow-to-deep training for neural ma- chine translation.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Shallow-to-deep training for neural ma- chine translation

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.754764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39a1bcaa-cc6c-41a5-8c03-bec1ba88d064 · outbound

This paper cites An image is worth 16x16 words: Transformers for image recognition at scale.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers An image is worth 16x16 words: Transformers for image recognition at scale

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:51.994756Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b20b22d-d874-4795-bebe-a0d7a08359e0 · outbound

This paper cites NormFormer: Improved Transformer Pretraining with Extra Normalization.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers NormFormer: Improved Transformer Pretraining with Extra Normalization

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:53.244800Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5565be61-45ba-488c-a31d-e5f4c4449a3b · outbound

This paper cites Transformers Get Stable: An End-to-End Signal Propagation Theory for Language Models.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Transformers Get Stable: An End-to-End Signal Propagation Theory for Language Models

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.424760Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 06bcf884-d9c7-4e6d-947d-a84f125f271c · outbound

This paper cites Peri-LN: Revisiting Normalization Layer in the Transformer Architecture.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Peri-LN: Revisiting Normalization Layer in the Transformer Architecture

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.586372Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3cfb67d6-0ef5-4585-a171-c128c013ec0c · outbound

This paper cites Mistral 7B.

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers Mistral 7B

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-03T06:37:52.334759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation c4654426-a0e4-448e-b38b-6af51e9993c0 · inbound

Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth cites this paper.

Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth SpanNorm: Reconciling Training Stability and Performance in Deep Transformers

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-02T03:08:35.290649Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74a1add0-7c06-4ea3-add0-8eff061bd9e7 · inbound

Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning cites this paper.

Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning SpanNorm: Reconciling Training Stability and Performance in Deep Transformers

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-01T15:58:13.171476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.