Paper Citation Record · LEDGER

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

As of 8 August 2026, this Paper Citation Record lists 44 of 44 outbound references and 2 inbound Pith citation observations for arXiv:2506.03850.

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

pith.paper-citation-record.v1
2506.03850 v2

Coverage vector

measured 44 of 44 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T10:58:32.629027Z

measured 46 of 46 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-06-30T21:01:25.549340Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-06-30T21:05:04.110492Z

Reference resolution

44 of 44 outbound references displayed

verified exact0
verified fuzzy14
unresolved30
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 85be4b79-7bc6-4a4b-8739-558370a5ff14 · outbound

This paper cites write newline.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:27.960888Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9201a311-3786-4f57-8a1a-d464c3807d6d · outbound

This paper cites an unresolved cited work.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Unresolved cited work

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:28.049879Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07745fbe-1cc1-4cbb-8cb0-60878b663edb · outbound

This paper cites D., Melenberg, B., and Rennen, G.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning D., Melenberg, B., and Rennen, G

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:35.806351Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 979dc913-5965-452a-ab2f-f847184f2920 · outbound

This paper cites Curriculum learning.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Curriculum learning

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:35.617713Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation fc7b2970-884a-458e-87cd-2e018560ef7b · outbound

This paper cites an unresolved cited work.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Unresolved cited work

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:28.349795Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b388d9c-4b5e-4e9b-a117-6132bfe03b9b · outbound

This paper cites Beyond factuality: A comprehensive evaluation of large language models as knowledge generators.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Beyond factuality: A comprehensive evaluation of large language models as knowledge generators

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:28.468902Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b99841c-5412-4628-8e93-984729d3a5e3 · outbound

This paper cites W at ME : Towards lossless watermarking through lexical redundancy.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning W at ME : Towards lossless watermarking through lexical redundancy

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:28.552869Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8d33798b-9766-4add-9d45-ff2fc5d2eb2d · outbound

This paper cites Simple permutations can fool LL a MA : Permutation attack and defense for large language models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Simple permutations can fool LL a MA : Permutation attack and defense for large language models

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:35.415083Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation c4a364ec-717d-4543-a634-bba1b2c0d6f6 · outbound

This paper cites PEARL : Towards permutation-resilient LLM s.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning PEARL : Towards permutation-resilient LLM s

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:35.239385Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation d4017aa4-c6e8-4a63-893e-37056372c498 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Training Verifiers to Solve Math Word Problems

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:28.922486Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ee26b7b9-f979-4f71-b552-d64461331b23 · outbound

This paper cites Statistics of robust optimization: A generalized empirical likelihood approach.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Statistics of robust optimization: A generalized empirical likelihood approach

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:35.054218Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 5b6e0b8d-9112-4c43-abf2-31217a9a4b2e · outbound

This paper cites Sharpness-aware minimization for efficiently improving generalization.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Sharpness-aware minimization for efficiently improving generalization

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:34.861219Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 0a7329e2-0b63-40cb-a845-8e5fd7ed970a · outbound

This paper cites A Survey of Uncertainty in Deep Neural Networks.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning A Survey of Uncertainty in Deep Neural Networks

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:29.226049Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d79474ea-dfc3-4cb8-9fc8-8c788cb0b3da · outbound

This paper cites Safe lora: the silver lining of reducing safety risks when fine-tuning large language models, 2024.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Safe lora: the silver lining of reducing safety risks when fine-tuning large language models, 2024

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:34.635342Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 3039b9af-f8d2-4aeb-b693-0f6cb6283dd8 · outbound

This paper cites Does distributionally robust supervised learning give robust classifiers? In International Conference on Machine Learning (ICML), 2018.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Does distributionally robust supervised learning give robust classifiers? In International Conference on Machine Learning (ICML), 2018

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:34.437632Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 5ce7a588-a395-4725-802b-23fa97cb6001 · outbound

This paper cites Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:29.556438Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 432e1bfa-79a9-4200-a759-08918ef0de8c · outbound

This paper cites Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:29.726809Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68912f69-44e1-4a26-9e9d-4eccb783f9cf · outbound

This paper cites Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:29.862212Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2b67a653-2ba4-40ac-b187-9c56cd80cf82 · outbound

This paper cites Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.014066Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85e40ba6-bc56-47e8-9791-7ac500928d3a · outbound

This paper cites Beavertails: Towards improved safety alignment of llm via a human-preference dataset.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Beavertails: Towards improved safety alignment of llm via a human-preference dataset

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.125758Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b62c561-52f3-42b0-9a61-828f87d4bb22 · outbound

This paper cites A watermark for large language models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning A watermark for large language models

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:34.258151Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 9045197d-47fa-47bb-aca2-3d50cea61a3d · outbound

This paper cites and Zhou, E.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning and Zhou, E

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:34.043488Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 992acd13-9279-4488-99b5-01a2cc5cbbb7 · outbound

This paper cites an unresolved cited work.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.489568Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56a9ff7c-d06a-43d1-a381-a3109563f274 · outbound

This paper cites Truthfulqa: Measuring how models mimic human falsehoods, 2022.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Truthfulqa: Measuring how models mimic human falsehoods, 2022

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.674810Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51fd9752-942b-42b2-90aa-7a1795e26c32 · outbound

This paper cites Decoupled Weight Decay Regularization.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Decoupled Weight Decay Regularization

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.814159Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4fed7e0f-c7a7-4eb8-808c-4691a40cde4a · outbound

This paper cites Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:30.941185Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 93941efd-55ef-4822-9be0-74563ea52630 · outbound

This paper cites Virtual adversarial training: a regularization method for supervised and semi-supervised learning.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Virtual adversarial training: a regularization method for supervised and semi-supervised learning

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:33.855444Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 8aca17d5-d9ff-4695-9543-791e580d92ad · outbound

This paper cites Fine-tuning can cripple your foundation model; preserving features may be the solution.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Fine-tuning can cripple your foundation model; preserving features may be the solution

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.154739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0899e5f3-95e0-4dfc-9c9b-61cd7d5dd285 · outbound

This paper cites Robust stochastic approximation approach to stochastic programming.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Robust stochastic approximation approach to stochastic programming

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.235844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 31bf7f29-8d3d-4195-b990-02fa3aec3dd0 · outbound

This paper cites Distributionally robust language modeling.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Distributionally robust language modeling

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:33.666994Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation ed815a8c-94fe-46df-8523-01939a05d727 · outbound

This paper cites Navigating the Safety Landscape: Measuring Risks in Finetuning Large Language Models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Navigating the Safety Landscape: Measuring Risks in Finetuning Large Language Models

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.384137Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9fa4308-f270-4c84-a1b8-eccdf39237e8 · outbound

This paper cites Gradient starvation: A learning proclivity in neural networks.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Gradient starvation: A learning proclivity in neural networks

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:33.450064Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 9213de61-e392-432a-957a-01f18cc978a4 · outbound

This paper cites Fine-tuning aligned language models compromises safety, even when users do not intend to! In The Twelfth International Conference on Learning Representations, 2024.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Fine-tuning aligned language models compromises safety, even when users do not intend to! In The Twelfth International Conference on Learning Representations, 2024

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:58:33.175737Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation e024dd50-b677-4397-9582-d93faf791fef · outbound

This paper cites Representation Noising: A Defence Mechanism Against Harmful Finetuning.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Representation Noising: A Defence Mechanism Against Harmful Finetuning

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.656919Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 692cf9e7-b23c-4523-93e3-a926ffd1e9cc · outbound

This paper cites Immunization against harmful fine-tuning attacks.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Immunization against harmful fine-tuning attacks

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.777027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6c52886-6a07-4020-a183-0c9c52698cd9 · outbound

This paper cites W., Hashimoto, T.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning W., Hashimoto, T

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.841598Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6970cdf6-317a-4e6d-be18-0ac31cd90c05 · outbound

This paper cites D., Ng, A.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning D., Ng, A

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:31.918620Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03d73a50-2a49-4b4e-841c-df5b920aea78 · outbound

This paper cites an unresolved cited work.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Unresolved cited work

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.018033Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 836a6827-2136-4cc3-99cc-2cc4c5eabf0b · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning LLaMA: Open and Efficient Foundation Language Models

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.116438Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4f4a639c-f73a-4dc8-adfe-3056e664ee1f · outbound

This paper cites Qwen2 Technical Report.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Qwen2 Technical Report

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.223942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e6a2e86-1dfb-49eb-bca1-21e2781a527e · outbound

This paper cites Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.316238Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 34a5e47a-548e-4a19-b784-3991868611b4 · outbound

This paper cites A safety realignment framework via subspace-oriented model fusion for large language models.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning A safety realignment framework via subspace-oriented model fusion for large language models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.433283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 382b390f-a3a8-4031-8574-31eb66cb9f27 · outbound

This paper cites Removing RLHF Protections in GPT-4 via Fine-Tuning.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Removing RLHF Protections in GPT-4 via Fine-Tuning

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.538591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c09f958c-6138-4837-b436-8b3cca594162 · outbound

This paper cites Character-level convolutional networks for text classification.

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning Character-level convolutional networks for text classification

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:32.629027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 7f15e16c-f8f7-4280-8880-d2d3f12f7f79 · inbound

One Step to the Side: Why Defenses Against Malicious Finetuning Fail Under Adaptive Adversaries cites this paper.

One Step to the Side: Why Defenses Against Malicious Finetuning Fail Under Adaptive Adversaries Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-06-30T21:05:04.112067Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 485d1a7f-70e7-45c9-800c-5887115546fb · inbound

SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection cites this paper.

SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-06-29T13:53:28.667357Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.