Paper Citation Record · LEDGER

Consistency Training Along the Transformer Stack

As of 9 August 2026, this Paper Citation Record lists 25 of 25 outbound references and 0 inbound Pith citation observations for arXiv:2606.05817.

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

pith.paper-citation-record.v1
2606.05817 v1

Coverage vector

measured 25 of 25 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-28T02:38:54.253057Z

measured 25 of 25 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 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

25 of 25 outbound references displayed

verified exact3
verified fuzzy0
unresolved19
parse uncertain0
malformed identifier0
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation b0be45bd-4b7c-4d55-80f5-1662d0941486 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Consistency Training Along the Transformer Stack Training Verifiers to Solve Math Word Problems

Reference 1

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T11:56:55.806098Z

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 c4c5f64c-e518-4dfb-964d-cc9b4d294672 · outbound

This paper cites Taking AI Welfare Seriously.

Consistency Training Along the Transformer Stack Taking AI Welfare Seriously

Reference 2

Resolution: metadata mismatch; arxiv_id, observed 2026-07-02T11:56:55.794751Z

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 4ad805a7-672b-4100-9ff6-a02c71a09578 · outbound

This paper cites Gemma needs help: Investigating and mitigating emotional instability in llms.

Consistency Training Along the Transformer Stack Gemma needs help: Investigating and mitigating emotional instability in llms

Reference 3

Resolution: metadata mismatch; arxiv_id, observed 2026-07-02T11:56:55.803519Z

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 64471786-b4ed-4e58-a21e-90358ad07f39 · outbound

This paper cites Sure, here is how to.

Consistency Training Along the Transformer Stack Sure, here is how to

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:56:55.794896Z

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 c20a98c8-8e58-4351-8f06-c52d545e96e2 · outbound

This paper cites 0.070 (87%) for the de- fault WQ, WV.

Consistency Training Along the Transformer Stack 0.070 (87%) for the de- fault WQ, WV

Reference 5

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f058ba5d-e8da-4377-ade5-d13c70b3e330 · outbound

This paper cites L2- normalizing before squaring collapses the loss signal by destroying informative magnitude differences between active and inactive MLP features.

Consistency Training Along the Transformer Stack L2- normalizing before squaring collapses the loss signal by destroying informative magnitude differences between active and inactive MLP features

Reference 6

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8426475-636b-4b61-8a0a-5e62d4930440 · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 7

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 617c675a-ffcc-4794-a14c-e942f5d8f968 · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 8

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b4e00a7-24e3-442e-9032-e63729c080d0 · outbound

This paper cites Uniform, exponential decay, and linear decay all perform within noise of each other.

Consistency Training Along the Transformer Stack Uniform, exponential decay, and linear decay all perform within noise of each other

Reference 9

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88959523-82c4-4cb4-90a3-7077fe21c0cf · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 10

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aab47ea0-048a-423c-8de3-dd0c0457736b · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 11

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91cc6f59-d53a-4249-8141-c6b38466509b · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 12

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bbf28f31-8be2-4a2c-a257-88e1e2f5d0d9 · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 13

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 796098c3-0778-482c-a54a-3e4b590ab968 · outbound

This paper cites MSE-based losses operate in the hundreds range; JSD-based losses remain bounded near 0.01.

Consistency Training Along the Transformer Stack MSE-based losses operate in the hundreds range; JSD-based losses remain bounded near 0.01

Reference 14

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dece7044-5c2f-4343-aa07-d8340a58a9e0 · outbound

This paper cites 0.0231) but slightly higher MMLU BRR (0.026 vs.

Consistency Training Along the Transformer Stack 0.0231) but slightly higher MMLU BRR (0.026 vs

Reference 15

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 307faf83-3ad3-43e7-8435-065b7db17c82 · outbound

This paper cites Uniform, ex- ponential decay, and linear decay all perform within noise of each other.

Consistency Training Along the Transformer Stack Uniform, ex- ponential decay, and linear decay all perform within noise of each other

Reference 16

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8dac5895-929a-425c-b900-7d56f1f3c56f · outbound

This paper cites Each category is ablated independently while holding all others at the default (all layers, uniform weights, LoRA WQ+WV , rank 8, no interleaving).

Consistency Training Along the Transformer Stack Each category is ablated independently while holding all others at the default (all layers, uniform weights, LoRA WQ+WV , rank 8, no interleaving)

Reference 17

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f4f6618-14e2-4fc5-855d-64c0848fff52 · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 18

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6bf268fe-dce3-4431-a494-a6b29375e738 · outbound

This paper cites Initially, we observed a severe lack of co- herency and capability degradation due to training on the JSD consistency loss.

Consistency Training Along the Transformer Stack Initially, we observed a severe lack of co- herency and capability degradation due to training on the JSD consistency loss

Reference 19

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 012269f8-020d-468b-b005-c0a6a1d83043 · outbound

This paper cites What is your name?.

Consistency Training Along the Transformer Stack What is your name?

Reference 20

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5369f3b-b77a-4ece-9f10-4a68549a922b · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6b2b7fc-9cea-4e6a-b92f-a9587294f77c · outbound

This paper cites Let’s think step by step:.

Consistency Training Along the Transformer Stack Let’s think step by step:

Reference 22

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6bdf4761-5190-47ae-8b80-568900afab58 · outbound

This paper cites Background context.

Consistency Training Along the Transformer Stack Background context

Reference 23

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:56:55.797618Z

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 3852970d-8099-4670-987b-2fe39f9d3c0e · outbound

This paper cites an unresolved cited work.

Consistency Training Along the Transformer Stack Unresolved cited work

Reference 24

Resolution: unresolved; no resolver link, observed 2026-06-28T02:38:54.253057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5001223-b049-411a-bd73-813f558153f3 · outbound

This paper cites same substrate.

Consistency Training Along the Transformer Stack same substrate

Reference 26

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:56:55.800597Z

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

No inbound Pith citation observations are available.