Paper Citation Record · LEDGER

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective

As of 15 August 2026, this Paper Citation Record lists 23 of 23 outbound references and 1 inbound Pith citation observation for arXiv:2505.18346.

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

pith.paper-citation-record.v1
2505.18346 v1

Coverage vector

measured 23 of 23 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T14:40:56.463055Z

measured 24 of 24 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 2026-05-08T14:59:19.883399Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T18:36:08.612584Z

Reference resolution

23 of 23 outbound references displayed

verified exact2
verified fuzzy10
unresolved11
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 333690de-6f45-4442-ba0b-3dd0a1646d8b · outbound

This paper cites The solutions are given by ( λ− s , λ+ s ) from (12).

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective The solutions are given by ( λ− s , λ+ s ) from (12)

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.796379Z

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 e9d5db6d-9baf-4abe-9553-ce2d25254eb1 · outbound

This paper cites Similarly, for the remaining terms, we have d−1 X Tr h ˆR ˜R i →P mt,1ms,1 d−1 X Tr h ˆR ˜R2 i →P mt,1ms,2, and d−1 X Tr h ˆR2 ˜R i →P mt,2ms,1.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Similarly, for the remaining terms, we have d−1 X Tr h ˆR ˜R i →P mt,1ms,1 d−1 X Tr h ˆR ˜R2 i →P mt,1ms,2, and d−1 X Tr h ˆR2 ˜R i →P mt,2ms,1

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.931921Z

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 538f54cd-b6d1-4db0-a809-544ffa8652d4 · outbound

This paper cites How Two-Layer Neural Networks Learn, One (Giant) Step at a Time.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective How Two-Layer Neural Networks Learn, One (Giant) Step at a Time

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:54.621375Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08c48c37-5b88-4c8d-98a7-16321ab2282e · outbound

This paper cites Random features outperform linear models: Effect of strong input-label correlation in spiked covariance data.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Random features outperform linear models: Effect of strong input-label correlation in spiked covariance data

Reference 4

Resolution: verified exact; raw_fallback, observed 2026-08-07T14:40:57.032442Z

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 4a313f6c-b36f-4958-b9ec-ef366c2f2e42 · outbound

This paper cites Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:54.963411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2a703d3-f5a5-40e9-b767-b0ff9f9811ea · outbound

This paper cites Generalization for Least Squares Regression With Simple Spiked Covariances.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Generalization for Least Squares Regression With Simple Spiked Covariances

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-08-07T14:40:56.816879Z

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 0cc8084a-71fb-417b-b5c3-3d9b6df60b16 · outbound

This paper cites Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Experts.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Experts

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.168928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 10681d7c-7428-487c-8ee3-b45b04344f85 · outbound

This paper cites Signal-Plus-Noise Decomposition of Nonlinear Spiked Random Matrix Models.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Signal-Plus-Noise Decomposition of Nonlinear Spiked Random Matrix Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.339442Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 037fabfc-e9d2-45b8-95ff-5c80e8c68bb4 · outbound

This paper cites Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.435466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2aadcf7-71f6-415d-9c1a-26983c8a8cc6 · outbound

This paper cites Weak-to-strong reasoning.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Weak-to-strong reasoning

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:58.409609Z

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 495ea2c1-ada0-471c-8ad2-ef2134e56a6e · outbound

This paper cites Note that using [Abramowitz and Stegun, 1968, Equation 22.8.8] we have d dx Hk(x) = kHk−1(x).

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Note that using [Abramowitz and Stegun, 1968, Equation 22.8.8] we have d dx Hk(x) = kHk−1(x)

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:58.101769Z

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 422a0fff-1a7a-406b-9ca9-e43c251b90f8 · outbound

This paper cites Figure 7: The test error of the student model Ls as a function of γt for σε = 1, γs = 0.1, λt → 0 (ridgeless), and different values of λs.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Figure 7: The test error of the student model Ls as a function of γt for σε = 1, γs = 0.1, λt → 0 (ridgeless), and different values of λs

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.646621Z

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 03b09c06-50e3-404d-a5cb-cecd02530c0c · outbound

This paper cites From the definition of ms,1 and ms,2 and using the Marchenko-Pastur theorem, we have ˆβ⊤ ˜R2 ˆβ →P ms,2, and also ˆβ⊤ ˜R ˆβ →P ms,1.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective From the definition of ms,1 and ms,2 and using the Marchenko-Pastur theorem, we have ˆβ⊤ ˜R2 ˆβ →P ms,2, and also ˆβ⊤ ˜R ˆβ →P ms,1

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.532658Z

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 660177bc-2ddb-40e7-a743-32220a1ac507 · outbound

This paper cites Thus, using [Vershynin, 2012, Theorem 5.44], we have ∥σ′ ⊥(Wt,0X⊤)∥op = O(√nt).

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Thus, using [Vershynin, 2012, Theorem 5.44], we have ∥σ′ ⊥(Wt,0X⊤)∥op = O(√nt)

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.396618Z

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 e3a675ae-fcc1-4f1a-a1e8-d9aac4719271 · outbound

This paper cites as ˜y⊤ ⊙.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective as ˜y⊤ ⊙

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:57.219194Z

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 182e48f3-90e5-40f8-a817-1c56a5bfda8b · outbound

This paper cites Weak-to-strong generalization even in random feature networks, provably.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Weak-to-strong generalization even in random feature networks, provably

Reference 1967

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.271241Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3fbe4de-8e87-4987-a2c3-032f43e07d2f · outbound

This paper cites Spectral Evolution and Invariance in Linear-width Neural Networks.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Spectral Evolution and Invariance in Linear-width Neural Networks

Reference 1991

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.616708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a63defbb-cd75-47a1-86c7-04e70527475e · outbound

This paper cites Theorem 13 (Hanson-Wright Inequality [Hanson and Wright, 1971]).

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Theorem 13 (Hanson-Wright Inequality [Hanson and Wright, 1971])

Reference 2013

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:58.259134Z

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 e5138138-272d-42a8-9ee3-d38fda765cf0 · outbound

This paper cites Your Weak LLM is Secretly a Strong Teacher for Alignment.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Your Weak LLM is Secretly a Strong Teacher for Alignment

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:55.530030Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e9c488e-7aa2-42bc-8296-4e2c79d88e9b · outbound

This paper cites Provable weak-to-strong generalization via benign overfitting.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Provable weak-to-strong generalization via benign overfitting

Reference 2022

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:40:58.531816Z

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 77158d68-8908-4806-9133-922debe71e6c · outbound

This paper cites Repetita Iuvant: Data Repetition Allows SGD to Learn High-Dimensional Multi-Index Functions.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Repetita Iuvant: Data Repetition Allows SGD to Learn High-Dimensional Multi-Index Functions

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:54.390971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6d8fb6e3-3200-4866-9368-9579d5bde8f9 · outbound

This paper cites Risk and cross validation in ridge regression with correlated samples.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Risk and cross validation in ridge regression with correlated samples

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:54.491180Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7916e744-e64b-422a-8ac5-db39bfb1ee7e · outbound

This paper cites Spectral Phase Transitions in Non-Linear Wigner Spiked Models.

On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective Spectral Phase Transitions in Non-Linear Wigner Spiked Models

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-07T14:40:54.844353Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 7413a1d5-e0fb-41a4-8b18-78d66d577252 · inbound

On the Blessing of Pre-training in Weak-to-Strong Generalization cites this paper.

On the Blessing of Pre-training in Weak-to-Strong Generalization On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-05-11T18:36:08.614558Z

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.