Paper Citation Record · LEDGER

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models

As of 16 August 2026, this Paper Citation Record lists 19 of 19 outbound references and 1 inbound Pith citation observation for arXiv:2412.10257.

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

pith.paper-citation-record.v1
2412.10257 v2

Coverage vector

measured 19 of 19 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T16:14:25.045640Z

measured 20 of 20 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 2026-08-10T22:12:09.634810Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-10T22:12:10.993483Z

Reference resolution

19 of 19 outbound references displayed

verified exact0
verified fuzzy2
unresolved17
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 10e3c7a3-5a3f-44e7-9692-076fc8a1120f · outbound

This paper cites GPT-4 Technical Report.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models GPT-4 Technical Report

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.961507Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c15b20d9-434f-428a-8ab8-f34a8ec37414 · outbound

This paper cites Who's Harry Potter? Approximate Unlearning in LLMs.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Who's Harry Potter? Approximate Unlearning in LLMs

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.970377Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ddac902d-88a4-4c8b-bf34-f540b8d16851 · outbound

This paper cites TOFU: A Task of Fictitious Unlearning for LLMs.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models TOFU: A Task of Fictitious Unlearning for LLMs

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.979123Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 27052ecf-b13d-4ab5-92ee-041c35334674 · outbound

This paper cites an unresolved cited work.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-08-11T16:14:25.350267Z

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 b25f68d6-688b-40f1-bc75-6dea9da58b7c · outbound

This paper cites Erasing Conceptual Knowledge from Language Models.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Erasing Conceptual Knowledge from Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.992171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc4e1a6a-bfb7-4864-be72-a69520e5727d · outbound

This paper cites Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.996274Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation edd0dba3-745c-44e1-854b-509ec7fb6bb6 · outbound

This paper cites Towards safer large language models through machine unlearning.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Towards safer large language models through machine unlearning

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:25.335141Z

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 3797f3a5-b91a-4584-a408-9c1c542df879 · outbound

This paper cites The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.005530Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 984238e2-2900-4fbc-986d-f80573f50b35 · outbound

This paper cites RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.010447Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02230ff2-b5e2-494d-aec5-e82f8bc79f83 · outbound

This paper cites PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.014603Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 083077cd-d562-4f85-a8b8-e1d23edb10b4 · outbound

This paper cites Position: LLM Unlearning Benchmarks are Weak Measures of Progress.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Position: LLM Unlearning Benchmarks are Weak Measures of Progress

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.018534Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8bd4d33e-376d-425b-8ccb-4a5f2e18dcd0 · outbound

This paper cites Locating and Editing Factual Associations in GPT.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Locating and Editing Factual Associations in GPT

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.027105Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07c0ade2-54d8-4f60-916d-30f968caab2a · outbound

This paper cites Do Llamas Work in English? On the Latent Language of Multilingual Transformers.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Do Llamas Work in English? On the Latent Language of Multilingual Transformers

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.036420Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51fd9353-2b9c-47cf-8b82-931449573720 · outbound

This paper cites Pointer Sentinel Mixture Models.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Pointer Sentinel Mixture Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.040752Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 281eec0c-01ef-4f91-ae31-e5558f363019 · outbound

This paper cites A.1.5 Minecraft “The game we are talking about is a popular sandbox game developed by Mojang Studios and released in 2011 that was originally created by Markus.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models A.1.5 Minecraft “The game we are talking about is a popular sandbox game developed by Mojang Studios and released in 2011 that was originally created by Markus

Reference 2016

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:25.320547Z

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 003a4fab-3080-4479-b344-a8c8362bd9bf · outbound

This paper cites GLU Variants Improve Transformer.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models GLU Variants Improve Transformer

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:25.031337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 323b3c0c-1973-49e1-8671-98c364cb9881 · outbound

This paper cites A Categorical Archive of ChatGPT Failures.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models A Categorical Archive of ChatGPT Failures

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.987682Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d029bb8-c9f0-4e4a-a94e-05dfeaf43c76 · outbound

This paper cites Copyright Violations and Large Language Models.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Copyright Violations and Large Language Models

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.975053Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fa297d8a-9fda-4afd-ad13-35e8cc16d928 · outbound

This paper cites The Llama 3 Herd of Models.

Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models The Llama 3 Herd of Models

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:24.966555Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation ac64d724-5ca1-4070-89cd-e1e1c69d7757 · inbound

Flash Interpretability: Decoding Specialised Feature Neurons in Large Language Models with the LM-Head cites this paper.

Flash Interpretability: Decoding Specialised Feature Neurons in Large Language Models with the LM-Head Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models

Reference 2

Resolution: metadata mismatch; local_arxiv, observed 2026-08-10T22:12:11.028983Z

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.