Paper Citation Record · LEDGER

MAGIC: Near-Optimal Data Attribution for Deep Learning

As of 22 August 2026, this Paper Citation Record lists 21 of 21 outbound references and 10 inbound Pith citation observations for arXiv:2504.16430.

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

pith.paper-citation-record.v1
2504.16430 v1

Coverage vector

measured 21 of 21 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T11:08:47.464823Z

measured 31 of 31 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 10 of 10 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-11T00:35:48.407645Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-02T16:17:08.354831Z

Reference resolution

21 of 21 outbound references displayed

verified exact2
verified fuzzy5
unresolved14
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 77e482bb-db5d-46d4-b11c-e29867d8c0a4 · outbound

This paper cites Training Data Attribution via Approximate Unrolled Differentiation.

MAGIC: Near-Optimal Data Attribution for Deep Learning Training Data Attribution via Approximate Unrolled Differentiation

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:46.903859Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef6cf447-ea5f-4cd5-83e1-1fee037b6fa6 · outbound

This paper cites Simfluence: Modeling the Influence of Individual Training Examples by Simulating Training Runs.

MAGIC: Near-Optimal Data Attribution for Deep Learning Simfluence: Modeling the Influence of Individual Training Examples by Simulating Training Runs

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:46.978734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29937b6e-cd5e-4665-8b45-e4cf9c5fc25e · outbound

This paper cites Training Data Influence Analysis and Estimation: A Survey.

MAGIC: Near-Optimal Data Attribution for Deep Learning Training Data Influence Analysis and Estimation: A Survey

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.055055Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e18f2c1-3071-4a02-8f95-0b1844cb4576 · outbound

This paper cites LoRA: Low-Rank Adaptation of Large Language Models.

MAGIC: Near-Optimal Data Attribution for Deep Learning LoRA: Low-Rank Adaptation of Large Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.104662Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b07c179a-c249-4b83-857a-3a8d979dd743 · outbound

This paper cites 94 percent on CIFAR-10 in 3.29 Secon ds on a Single GPU.

MAGIC: Near-Optimal Data Attribution for Deep Learning 94 percent on CIFAR-10 in 3.29 Secon ds on a Single GPU

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:08:47.968733Z

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 94046e50-e258-4b4c-a1cf-21f323b5c526 · outbound

This paper cites Generalized Group Data Attribution.

MAGIC: Near-Optimal Data Attribution for Deep Learning Generalized Group Data Attribution

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.316990Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5fa77785-7bfc-4a07-b664-a524f8641202 · outbound

This paper cites Optimizing millions of hyper- parameters by implicit differentiation.

MAGIC: Near-Optimal Data Attribution for Deep Learning Optimizing millions of hyper- parameters by implicit differentiation

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:08:47.830948Z

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 4f2bc213-ddda-4a23-a9d4-9fc42694ee0b · outbound

This paper cites RandALO: Out-of-sample risk estimation in no time flat.

MAGIC: Near-Optimal Data Attribution for Deep Learning RandALO: Out-of-sample risk estimation in no time flat

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-08-16T11:08:47.584966Z

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 1fa7c193-51ac-4ed4-a9de-73f5f8a6ad9b · outbound

This paper cites TRAK: Attributing Model Behavior at Scale.

MAGIC: Near-Optimal Data Attribution for Deep Learning TRAK: Attributing Model Behavior at Scale

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.330040Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66df48f2-da0a-43a8-9b68-2dee36d20cb2 · outbound

This paper cites Learning Transferable Visual Models From Natural Language Supervision.

MAGIC: Near-Optimal Data Attribution for Deep Learning Learning Transferable Visual Models From Natural Language Supervision

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.446156Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e86947d9-341c-47c0-9446-13bfd3f1e18c · outbound

This paper cites A scalable estimate of the extra-sample prediction error via approximate leave-one-out.

MAGIC: Near-Optimal Data Attribution for Deep Learning A scalable estimate of the extra-sample prediction error via approximate leave-one-out

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.450519Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95929129-17c8-468f-9a1d-d244e5c131a1 · outbound

This paper cites Language models are unsupervised multitask learner s.

MAGIC: Near-Optimal Data Attribution for Deep Learning Language models are unsupervised multitask learner s

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:08:47.818786Z

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 a6585843-fc01-4666-aeed-8986b0689449 · outbound

This paper cites ModelDiff: A Framework for Comparing Learning Algorithms.

MAGIC: Near-Optimal Data Attribution for Deep Learning ModelDiff: A Framework for Comparing Learning Algorithms

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.460064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1be4f03-ae42-44a4-bc4b-c8ed838a42f3 · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

MAGIC: Near-Optimal Data Attribution for Deep Learning Gemma: Open Models Based on Gemini Research and Technology

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.464823Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 819c47fb-a3d8-4f46-bea1-daa6a28661a5 · outbound

This paper cites Scaling Laws for Neural Language Models.

MAGIC: Near-Optimal Data Attribution for Deep Learning Scaling Laws for Neural Language Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:47.241135Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e772eb1b-f0dd-43b2-b088-49ee1dfcfd81 · outbound

This paper cites Contrastive Error Attribution for Finetuned Language Models.

MAGIC: Near-Optimal Data Attribution for Deep Learning Contrastive Error Attribution for Finetuned Language Models

Reference 2018

Resolution: verified exact; local_arxiv, observed 2026-08-16T11:08:47.710679Z

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 4e62f9d0-005d-49fa-9e4e-635f0c14c879 · outbound

This paper cites Openassistant conversations-democratizing large la nguage model alignment.

MAGIC: Near-Optimal Data Attribution for Deep Learning Openassistant conversations-democratizing large la nguage model alignment

Reference 2019

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:08:47.843494Z

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 2445795c-e809-4ab0-a15b-0105656d3f78 · outbound

This paper cites Mod el multiplicity: Opportuni- ties, concerns, and solutions.

MAGIC: Near-Optimal Data Attribution for Deep Learning Mod el multiplicity: Opportuni- ties, concerns, and solutions

Reference 2021

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:08:47.980242Z

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 07f5071c-5237-4558-9bd4-d916bdbd0f5e · outbound

This paper cites Studying Large Language Model Generalization with Influence Functions.

MAGIC: Near-Optimal Data Attribution for Deep Learning Studying Large Language Model Generalization with Influence Functions

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:46.973544Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6347dc68-e27d-42f9-ad83-2f6ac21c54a4 · outbound

This paper cites Optimizing ML Training with Metagradient Descent.

MAGIC: Near-Optimal Data Attribution for Deep Learning Optimizing ML Training with Metagradient Descent

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:46.969452Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 16d63ea9-94f1-4130-a1d1-1bf9432c0b6d · outbound

This paper cites If Influence Functions are the Answer, Then What is the Question?.

MAGIC: Near-Optimal Data Attribution for Deep Learning If Influence Functions are the Answer, Then What is the Question?

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-16T11:08:46.961415Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 4c7db193-a0bf-4e0e-85a7-89e3682e91f4 · inbound

Better Training Data Attribution via Better Inverse Hessian-Vector Products cites this paper.

Better Training Data Attribution via Better Inverse Hessian-Vector Products MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-06T15:54:42.424199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 589b7f10-d0f3-4193-a0e8-645a5b32c470 · inbound

LLM generation novelty through the lens of semantic similarity cites this paper.

LLM generation novelty through the lens of semantic similarity MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-04T07:22:05.367970Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf6fc1fa-11cb-4836-a304-ed9e132e9fde · inbound

Efficient Estimation of Kernel Surrogate Models for Task Attribution cites this paper.

Efficient Estimation of Kernel Surrogate Models for Task Attribution MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 5

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T08:00:44.858296Z

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 ece4c010-1723-45db-b8eb-2aad9ad3324b · inbound

How to sketch a learning algorithm cites this paper.

How to sketch a learning algorithm MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 10

Resolution: metadata mismatch; arxiv_id, observed 2026-05-11T05:30:56.453850Z

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 11c75c0c-7f29-4b56-afba-2bdccf6d5149 · inbound

How Faithful Is Trajectory-Based Data Attribution? Error Sources, Remedies, and Practical Guidelines cites this paper.

How Faithful Is Trajectory-Based Data Attribution? Error Sources, Remedies, and Practical Guidelines MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-20T23:13:50.446204Z

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 0877e611-a7e8-44be-a61b-f9f97bcfde6f · inbound

Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamics cites this paper.

Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamics MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-07-02T07:56:47.951304Z

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 2ac5f64b-8551-47a0-9bdc-da75901720b0 · inbound

Small edits, large models: How Wikipedia advocacy shapes LLM values cites this paper.

Small edits, large models: How Wikipedia advocacy shapes LLM values MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 8

Resolution: metadata mismatch; arxiv_id, observed 2026-07-01T09:05:36.485562Z

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 56ead02d-163d-4a52-9e1a-53f1a5415b7b · inbound

Small edits, large models: How Wikipedia advocacy shapes LLM values cites this paper.

Small edits, large models: How Wikipedia advocacy shapes LLM values MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-07-14T19:23:22.815333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 24ec7b89-af0a-4242-aa7d-bde8c7fb848c · inbound

Prototype Language Models cites this paper.

Prototype Language Models MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 163

Resolution: metadata mismatch; arxiv_id, observed 2026-07-02T16:17:08.356434Z

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 a3d87da3-04e9-4c0d-a488-405dc6c7736c · inbound

Tracing sources of epistemic uncertainty in deep learning predictions: homo- and hetero-scedastic linearized estimators cites this paper.

Tracing sources of epistemic uncertainty in deep learning predictions: homo- and hetero-scedastic linearized estimators MAGIC: Near-Optimal Data Attribution for Deep Learning

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-11T00:35:48.407645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.