Pith. sign in

REVIEW 3 cited by

MAGIC: Near-Optimal Data Attribution for Deep Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.16430 v1 pith:MQLF6WX2 submitted 2025-04-23 cs.LG cs.CLcs.CVstat.ML

classification cs.LGcs.CLcs.CVstat.ML
keywords dataattributionmethodsaddingestimategoalmagicmodel
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The goal of predictive data attribution is to estimate how adding or removing a given set of training datapoints will affect model predictions. In convex settings, this goal is straightforward (i.e., via the infinitesimal jackknife). In large-scale (non-convex) settings, however, existing methods are far less successful -- current methods' estimates often only weakly correlate with ground truth. In this work, we present a new data attribution method (MAGIC) that combines classical methods and recent advances in metadifferentiation to (nearly) optimally estimate the effect of adding or removing training data on model predictions.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Small edits, large models: How Wikipedia advocacy shapes LLM values

    cs.CL 2026-04 unverdicted novelty 7.0 of 10

    Wikipedia edits by animal welfare advocates measurably influence LLM outputs on animal welfare topics, shown via retrieval and gradient attribution plus fine-tuning experiments.

  2. LLM generation novelty through the lens of semantic similarity

    cs.LG 2025-10 conditional novelty 6.0 of 10

    An output is called novel when no semantically similar pretraining context can be retrieved; using this embedding-based measure, SmolLM models show longer-range reuse and instruction tuning increases novelty.

  3. Better Training Data Attribution via Better Inverse Hessian-Vector Products

    cs.LG 2025-07 conditional novelty 6.0 of 10

    ASTRA, an EKFAC-preconditioned Neumann series iteration, computes more accurate inverse Hessian-vector products and improves training data attribution scores over EKFAC baselines.

Pith tools