Paper Citation Record · LEDGER

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning

As of 20 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 0 inbound Pith citation observations for arXiv:2607.21353.

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

pith.paper-citation-record.v1
2607.21353 v1

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-01T07:47:12.190798Z

measured 24 of 24 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-20T06:33:59.587034+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

24 of 24 outbound references displayed

verified exact1
verified fuzzy0
unresolved23
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation c890e1fd-8239-49ec-9d35-9e8de3271979 · outbound

This paper cites the mask does nothing.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning the mask does nothing

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:12.033610Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a616efc-e48b-40ad-9802-7a40371d18bd · outbound

This paper cites An Illusion of Unlearning? Assessing Machine Unlearning Through Internal Representations.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning An Illusion of Unlearning? Assessing Machine Unlearning Through Internal Representations

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.783404Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e4e3999-ea26-493a-89e3-8be18532f7ba · outbound

This paper cites Corrective Machine Unlearning.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Corrective Machine Unlearning

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.949973Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9db907eb-e647-459f-b5cd-4774d16e6201 · outbound

This paper cites Gone but Not Forgotten: Improved Benchmarks for Machine Unlearning.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Gone but Not Forgotten: Improved Benchmarks for Machine Unlearning

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.050953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0c3d86e1-9e2b-402d-a61d-e8512b6c3b3a · outbound

This paper cites gradient concentration, not weight saliency, explains representation-level class unlearning.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning gradient concentration, not weight saliency, explains representation-level class unlearning

Reference 9

Resolution: verified exact; doi, observed 2026-08-01T07:48:30.402380Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation f2faac91-81b3-4825-97e0-010f9c7120c9 · outbound

This paper cites Class Unlearning via Depth-Aware Removal of Forget-Specific Directions.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Class Unlearning via Depth-Aware Removal of Forget-Specific Directions

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.412901Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff2c1f8e-febd-4eaa-9d22-1162a397a426 · outbound

This paper cites Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.473093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c96d217-b954-4f7d-8309-d7ada357a5a7 · outbound

This paper cites One-Point Contraction: Erasing Representational Separability toward Irreversible Deep Forgetting.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning One-Point Contraction: Erasing Representational Separability toward Irreversible Deep Forgetting

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.525403Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 19851a51-fa71-446a-bf2a-5093bae4a4b1 · outbound

This paper cites arXiv preprint arXiv:2503.06991.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning arXiv preprint arXiv:2503.06991

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.568328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 93f88a65-2018-4b9b-97c4-286e4147dd17 · outbound

This paper cites POUR: A Provably Optimal Method for Unlearning Representations via Neural Collapse.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning POUR: A Provably Optimal Method for Unlearning Representations via Neural Collapse

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.611722Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db0331c2-1ed5-4234-aa9e-5f214f472986 · outbound

This paper cites arXiv preprint arXiv:2602.05375.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning arXiv preprint arXiv:2602.05375

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.668475Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb0a8d0e-46a0-4aae-8e67-f5545104aa69 · outbound

This paper cites A Comprehensive Survey of Machine Unlearning Techniques for Large Language Models.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning A Comprehensive Survey of Machine Unlearning Techniques for Large Language Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.800199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04950356-b81b-4b3a-a91f-8eccbf16b081 · outbound

This paper cites A Survey of Machine Unlearning.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning A Survey of Machine Unlearning

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.843091Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08f6c48f-6655-4f69-8070-115254fdcaa5 · outbound

This paper cites arXiv preprint arXiv:2406.17216.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning arXiv preprint arXiv:2406.17216

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.886464Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 264052c7-26be-4148-bf05-6ebfac558cf7 · outbound

This paper cites arXiv preprint arXiv:2601.22028.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning arXiv preprint arXiv:2601.22028

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.929949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 669921da-25f8-4b52-a5f8-68433bf18862 · outbound

This paper cites Erased, but Not Gone: Output Forgetting Is Not True Forgetting.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Erased, but Not Gone: Output Forgetting Is Not True Forgetting

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.986123Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ca6bbd7-f845-4bbc-983e-851c7c4b48c8 · outbound

This paper cites an unresolved cited work.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Unresolved cited work

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:12.088587Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 195e9a9b-7f89-40c3-ad17-a0c3c8c2890c · outbound

This paper cites Ten-class means at the primary seed: Base0.66, Gold−0.09, SalUn0.62, RL-full0.68, Random0.67, Fisher0.61, Early0.56, Early3×0.53.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Ten-class means at the primary seed: Base0.66, Gold−0.09, SalUn0.62, RL-full0.68, Random0.67, Fisher0.61, Early0.56, Early3×0.53

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:12.137031Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd5a8011-9b45-4c12-a88c-2fb2e61050ae · outbound

This paper cites He haspublishedmorethan120scientificpapersininternational conferences and journals.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning He haspublishedmorethan120scientificpapersininternational conferences and journals

Reference 2010

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:12.190798Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7de84786-3e1a-4964-ac4d-f327938ff9e6 · outbound

This paper cites Making AI Forget You: Data Deletion in Machine Learning.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Making AI Forget You: Data Deletion in Machine Learning

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.891287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f65bda7d-57a3-420b-8b5e-1dca03d6529e · outbound

This paper cites 1897–1914.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning 1897–1914

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.645346Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a1bec359-e0d7-480b-a64e-40e6d9c443e7 · outbound

This paper cites SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.725197Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6996f491-b63b-47ec-b0c9-96037ef629f7 · outbound

This paper cites Unlearning's Blind Spots: Over-Unlearning and Prototypical Relearning Attack.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Unlearning's Blind Spots: Over-Unlearning and Prototypical Relearning Attack

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:11.267936Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c97ca099-d094-4236-b341-7afbec696f2d · outbound

This paper cites Representation Unlearning: Forgetting through Information Compression.

Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning Representation Unlearning: Forgetting through Information Compression

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-01T07:47:10.562679Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.