Paper Citation Record · LEDGER

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

As of 22 August 2026, this Paper Citation Record lists 92 of 92 outbound references and 9 inbound Pith citation observations for arXiv:2502.05209.

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

pith.paper-citation-record.v1
2502.05209 v4

Coverage vector

measured 92 of 92 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-09T14:47:15.448064Z

measured 101 of 101 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 9 of 9 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-15T16:25:35.442838Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-18T10:46:17.055047Z

Reference resolution

92 of 92 outbound references displayed

verified exact1
verified fuzzy13
unresolved78
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation bf1f4201-9b52-47eb-a0b6-d553ec611530 · outbound

This paper cites write newline.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:14.990982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 183e7ffa-d049-4cfb-b8cf-0c3e179c6892 · outbound

This paper cites T., O'Brien, J., Soder, L., Bucknall, B., Bluemke, E., Schuett, J., Trager, R., Strahm, L., and Chowdhury, R.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities T., O'Brien, J., Soder, L., Bucknall, B., Bluemke, E., Schuett, J., Trager, R., Strahm, L., and Chowdhury, R

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:14.997251Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73b1af65-f276-4185-8a02-8c85bf245625 · outbound

This paper cites Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.002017Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30f4dc67-561b-4b2f-a9fa-dbd1744c6fb6 · outbound

This paper cites Many-shot jailbreaking.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Many-shot jailbreaking

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.007519Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3e72b8b-62a8-481d-8271-b19d296bebaa · outbound

This paper cites Unlearning in large language models via activation projections, 2025.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unlearning in large language models via activation projections, 2025

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.012554Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b60765c1-c7ba-4511-b905-6f60d5639e29 · outbound

This paper cites Refusal in Language Models Is Mediated by a Single Direction.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Refusal in Language Models Is Mediated by a Single Direction

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.017432Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 633bc806-8686-4c0c-bb0e-16e5805b4fdd · outbound

This paper cites MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.023992Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9d4dc73-3dd6-4f1b-be7c-ac44238e7d1c · outbound

This paper cites Open Problems in Machine Unlearning for AI Safety.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Open Problems in Machine Unlearning for AI Safety

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.029821Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6fcfcd8-651e-4314-9848-00fa399ccd0d · outbound

This paper cites Language Model Unalignment: Parametric Red-Teaming to Expose Hidden Harms and Biases.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Language Model Unalignment: Parametric Red-Teaming to Expose Hidden Harms and Biases

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.034861Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4f9795a-a64c-442d-95fd-3ba18fbca45d · outbound

This paper cites an unresolved cited work.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unresolved cited work

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.040084Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a305ddd1-337a-4a51-a1d5-5426e98c3f80 · outbound

This paper cites AI and Data Act: Part of Bill C-27, Digital Charter Implementation Act, 2022 , 2022.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities AI and Data Act: Part of Bill C-27, Digital Charter Implementation Act, 2022 , 2022

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.044779Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9e2320bc-7954-45f0-8146-5a6b68311de4 · outbound

This paper cites A., Jagielski, M., Gao, I., Koh, P.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities A., Jagielski, M., Gao, I., Koh, P

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.049489Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 375996cf-fb2f-4c96-8acc-b63e4b626094 · outbound

This paper cites Defending Against Unforeseen Failure Modes with Latent Adversarial Training.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Defending Against Unforeseen Failure Modes with Latent Adversarial Training

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.054255Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b042a6f-9ed9-45de-ada7-14c98f4f361c · outbound

This paper cites Jailbreaking Black Box Large Language Models in Twenty Queries.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Jailbreaking Black Box Large Language Models in Twenty Queries

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.059261Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5b04b81-e9bb-4b32-bb5f-815e9bd6bf24 · outbound

This paper cites Interim Measures for the Management of Generative Artificial Intelligence Services , 2023.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Interim Measures for the Management of Generative Artificial Intelligence Services , 2023

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.064636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 037a7571-e26d-4a03-8e60-13c6c5568b60 · outbound

This paper cites G., Islam, M.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities G., Islam, M

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.069152Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9fd04bde-0107-41eb-b320-f5373135624a · outbound

This paper cites O., and Nilsson, F.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities O., and Nilsson, F

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.073705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f515d96-f150-4a66-87a0-e43671afe57a · outbound

This paper cites Do Unlearning Methods Remove Information from Language Model Weights?.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Do Unlearning Methods Remove Information from Language Model Weights?

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.078108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e625dda2-cebb-471c-a6ef-c43baa3c9637 · outbound

This paper cites The Llama 3 Herd of Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities The Llama 3 Herd of Models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.082764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fafdf18d-29b6-43e4-b8a6-8cf8436227fa · outbound

This paper cites The eu artificial intelligence act.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities The eu artificial intelligence act

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.087548Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation feea78e1-b98f-41dc-94a2-5ebc51bc4061 · outbound

This paper cites Scaling Laws for Adversarial Attacks on Language Model Activations.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Scaling Laws for Adversarial Attacks on Language Model Activations

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.092289Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b9d2a9f-2592-4aa0-9ac8-804cadcd21c5 · outbound

This paper cites Towards a science of ai evaluations.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Towards a science of ai evaluations

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.097049Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db1812b8-3cba-45e1-ab28-f5d46d864bf3 · outbound

This paper cites Erasing Conceptual Knowledge from Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Erasing Conceptual Knowledge from Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.101899Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef386264-db9e-4fb4-bfb3-572786246e45 · outbound

This paper cites Stress-Testing Capability Elicitation With Password-Locked Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Stress-Testing Capability Elicitation With Password-Locked Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.106879Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ced76700-4f74-427c-98f4-f4fd392bcd53 · outbound

This paper cites Cascade: Exploring hierarchical inference in language models, 2023.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Cascade: Exploring hierarchical inference in language models, 2023

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.111767Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 122f75cd-0c4e-4a4f-b5d9-096bb7b5eeee · outbound

This paper cites T., Haghtalab, N., and Steinhardt, J.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities T., Haghtalab, N., and Steinhardt, J

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.947064Z

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 859381dd-9d09-4cc8-ab4a-ecce32ab1439 · outbound

This paper cites Measuring Massive Multitask Language Understanding.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Measuring Massive Multitask Language Understanding

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.121434Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a1b4633c-ab77-48bb-be91-52cd0af80adc · outbound

This paper cites an unresolved cited work.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unresolved cited work

Reference 28

Resolution: unresolved; raw_fallback, observed 2026-08-09T14:47:35.928848Z

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 ac493279-4922-49e3-a797-b466f436aed1 · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities LoRA: Low-Rank Adaptation of Large Language Models

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.130612Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c267204e-d9f5-4a06-a2ef-70836602af39 · outbound

This paper cites Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.135860Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8dd47542-8160-46b8-a070-04a1f37247bd · outbound

This paper cites Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.141334Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3bad5977-5ee9-49b1-a799-7ab9661c9872 · outbound

This paper cites Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.146775Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 853c3ab5-e289-40be-89ea-4a61d4329711 · outbound

This paper cites Language models resist alignment, 2024.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Language models resist alignment, 2024

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.911752Z

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 beaf5ec8-3b43-4564-b970-5b3fa8a64530 · outbound

This paper cites Jailbreakzoo: Survey, landscapes, and horizons in jailbreaking large language and vision-language models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Jailbreakzoo: Survey, landscapes, and horizons in jailbreaking large language and vision-language models

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.157036Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c6eb7ce-c036-404c-b7b8-55c92326ee63 · outbound

This paper cites Act on the protection of personal information, 2025.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Act on the protection of personal information, 2025

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.894782Z

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 8bb16e20-1d1b-44ef-82f2-e26f100e351b · outbound

This paper cites No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.166472Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 43acf8e1-359d-47e2-b21a-ca64edf4fd8d · outbound

This paper cites LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.171392Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e85b16c9-7c66-4b34-93ef-ddfa98ce1114 · outbound

This paper cites LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.176057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6c86131-280a-44a2-b768-ca80477cde25 · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.181547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7569bbfc-337a-4e57-827e-c21e17128f0b · outbound

This paper cites Against The Achilles' Heel: A Survey on Red Teaming for Generative Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Against The Achilles' Heel: A Survey on Red Teaming for Generative Models

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.186818Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0c5d3e0-6632-47ea-a285-1a98c0d9d47c · outbound

This paper cites Continual learning and private unlearning.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Continual learning and private unlearning

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.878763Z

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 9a67d7ca-f77f-4652-8175-fc201462abfa · outbound

This paper cites Rethinking Machine Unlearning for Large Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Rethinking Machine Unlearning for Large Language Models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.196404Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 153ff993-0b35-4b9e-a6a0-c6be10f7c39e · outbound

This paper cites Threats, Attacks, and Defenses in Machine Unlearning: A Survey.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Threats, Attacks, and Defenses in Machine Unlearning: A Survey

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.201431Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dfa340c9-2760-493a-9f3e-6fed376a1db0 · outbound

This paper cites Large Language Models Relearn Removed Concepts.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Large Language Models Relearn Removed Concepts

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.206547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b1a68a4a-4299-4fca-ba6e-6f07c0521583 · outbound

This paper cites and Rimsky, N.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities and Rimsky, N

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.862145Z

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 52289a63-379a-44f5-b904-0db3406e363f · outbound

This paper cites An Adversarial Perspective on Machine Unlearning for AI Safety.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities An Adversarial Perspective on Machine Unlearning for AI Safety

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.215964Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c24c5f8c-9705-45b0-b204-b9a06d97a898 · outbound

This paper cites Eight Methods to Evaluate Robust Unlearning in LLMs.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Eight Methods to Evaluate Robust Unlearning in LLMs

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.221048Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ce66b5b-93f5-4eb5-883c-4f0784adca3a · outbound

This paper cites Pointer sentinel mixture models, 2016.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Pointer sentinel mixture models, 2016

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.225782Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a9b5cb3-561d-4abb-83f5-92739136e69a · outbound

This paper cites an unresolved cited work.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unresolved cited work

Reference 49

Resolution: unresolved; raw_fallback, observed 2026-08-09T14:47:35.833753Z

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 f24989a3-4478-4e69-aef5-a8658535b9d4 · outbound

This paper cites AI Risk Management Framework : AI RMF (1.0), January 2023.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities AI Risk Management Framework : AI RMF (1.0), January 2023

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.817088Z

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 94250e00-55ce-41fe-82d0-0e93f89049fe · outbound

This paper cites Openai system card: December 2024, 2024.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Openai system card: December 2024, 2024

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.799205Z

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 11943eb2-7684-42a4-9402-04f4fe61a795 · outbound

This paper cites Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.244674Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 45cc7030-3d76-4a04-a0b8-f5b5f5e44703 · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Navigating the Safety Landscape: Measuring Risks in Finetuning Large Language Models

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.249627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9921514-bedd-4a3d-a40f-c3ecc1d08f61 · outbound

This paper cites Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.254462Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7657da45-83b4-4452-96cc-aaf6b36efad7 · outbound

This paper cites Safety alignment should be made more than just a few tokens deep, 2024 a.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Safety alignment should be made more than just a few tokens deep, 2024 a

Reference 55

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.783616Z

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 46082899-8732-4da9-ad97-7c6104d9b44c · outbound

This paper cites On Evaluating the Durability of Safeguards for Open-Weight LLMs.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities On Evaluating the Durability of Safeguards for Open-Weight LLMs

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.264207Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 46e855c5-3c66-40d2-b412-a59fd126938b · outbound

This paper cites D., Xu, P., Honigsberg, C., and Ho, D.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities D., Xu, P., Honigsberg, C., and Ho, D

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.767226Z

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 02e2e5a7-928d-45c4-b7d0-34193dd1bf67 · outbound

This paper cites Open Problems in Technical AI Governance.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Open Problems in Technical AI Governance

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.273714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1c44e37-5f68-4494-8618-560ad26d944d · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Representation Noising: A Defence Mechanism Against Harmful Finetuning

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.278636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc0996cd-9a84-41e8-95b7-8e63d29f3799 · outbound

This paper cites Fast Adversarial Attacks on Language Models In One GPU Minute.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Fast Adversarial Attacks on Language Models In One GPU Minute

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.283768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 494f9dc6-20ee-48f4-825c-e56e317b936b · outbound

This paper cites an unresolved cited work.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unresolved cited work

Reference 61

Resolution: unresolved; raw_fallback, observed 2026-08-09T14:47:35.749567Z

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 77db4e86-0d51-49ef-b89a-be27da8b1815 · outbound

This paper cites Towards best practices in AGI safety and governance: A survey of expert opinion.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Towards best practices in AGI safety and governance: A survey of expert opinion

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.293676Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6ccc3f80-0418-4903-a57f-661b4c4fc24c · outbound

This paper cites Adversarial attacks and defenses in large language models: Old and new threats.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Adversarial attacks and defenses in large language models: Old and new threats

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.730884Z

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 ad40a32b-3f19-4bfc-b54c-e45792a8415a · outbound

This paper cites Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.304017Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation af0cb92b-f7bd-42ad-9c34-e3a1ece7f36f · outbound

This paper cites Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.309285Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b9b18c3-6020-455f-9cdc-8ab91d420ad9 · outbound

This paper cites Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.314358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e882aafa-ceab-4c62-9b51-c96884513ba0 · outbound

This paper cites Model evaluation for extreme risks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Model evaluation for extreme risks

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.319250Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04b13167-b951-403d-9833-4c2ac64427dc · outbound

This paper cites AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.324252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b1f053a-be81-4969-aa6a-d73c134f95ac · outbound

This paper cites UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.329073Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6df5fa0-756f-4281-87e5-1492cc8cc155 · outbound

This paper cites an unresolved cited work.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Unresolved cited work

Reference 70

Resolution: unresolved; raw_fallback, observed 2026-08-09T14:47:35.714557Z

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 3b01b132-ce61-46a2-aedf-4dcea52fdc81 · outbound

This paper cites A StrongREJECT for Empty Jailbreaks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities A StrongREJECT for Empty Jailbreaks

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.338954Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1d3ddeba-b616-47bb-bed5-613a2bf254ed · outbound

This paper cites A Simple and Effective Pruning Approach for Large Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities A Simple and Effective Pruning Approach for Large Language Models

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.343866Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a2486f0-3e1b-4f9f-b395-d26a1835bd4d · outbound

This paper cites Tamper-Resistant Safeguards for Open-Weight LLMs.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Tamper-Resistant Safeguards for Open-Weight LLMs

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.348837Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aa727472-dfef-42ec-84b1-83bb5e94f6d2 · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.353590Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1dfc3555-196d-425e-833a-f95e4ee86445 · outbound

This paper cites A pro-innovation approach to AI regulation.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities A pro-innovation approach to AI regulation

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.698068Z

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 1876fddd-203d-463e-839f-4af8685db792 · outbound

This paper cites AI Sandbagging: Language Models can Strategically Underperform on Evaluations.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities AI Sandbagging: Language Models can Strategically Underperform on Evaluations

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.363084Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd41a068-d301-48c4-a177-4c4270a48e03 · outbound

This paper cites Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.367750Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0bf8c76c-ecb1-4d01-a8c8-a7a07a00873c · outbound

This paper cites Efficient Adversarial Training in LLMs with Continuous Attacks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Efficient Adversarial Training in LLMs with Continuous Attacks

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.372451Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da6e0ecc-79bb-47cc-80ab-863b8edc129c · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.377658Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc9df143-d719-4af9-8b61-8e95f72f1ce9 · outbound

This paper cites On the vulnerability of safety alignment in open-access llms.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities On the vulnerability of safety alignment in open-access llms

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.680996Z

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 c0b52edb-205f-49c1-879f-be6937fc0b5e · outbound

This paper cites Jailbreak Attacks and Defenses Against Large Language Models: A Survey.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Jailbreak Attacks and Defenses Against Large Language Models: A Survey

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.387716Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc50e58e-37ee-492d-bc8c-52347c806179 · outbound

This paper cites Low-Resource Languages Jailbreak GPT-4.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Low-Resource Languages Jailbreak GPT-4

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.392539Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aefc68dd-f5a7-4218-8fe3-77f7c6f34395 · outbound

This paper cites Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.397309Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad4eb8ad-9be3-49ed-bbd1-b106c3b322d5 · outbound

This paper cites BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.402376Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75cfd0a4-16c1-4066-8ef6-5fb86992c319 · outbound

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

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Removing RLHF Protections in GPT-4 via Fine-Tuning

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.407456Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f3323a75-f9ad-4ce3-b497-b1bad5992b69 · outbound

This paper cites Adversarial machine learning in latent representations of neural networks.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Adversarial machine learning in latent representations of neural networks

Reference 86

Resolution: verified exact; arxiv_id, observed 2026-08-09T14:47:24.387209Z

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 73905ea9-a758-443f-8657-588ec377ee6c · outbound

This paper cites Catastrophic Failure of LLM Unlearning via Quantization.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Catastrophic Failure of LLM Unlearning via Quantization

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.422836Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c52482b9-2f28-4048-80f8-85433775e0de · outbound

This paper cites A Survey of Recent Backdoor Attacks and Defenses in Large Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities A Survey of Recent Backdoor Attacks and Defenses in Large Language Models

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.427626Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f86f630-1d5a-440f-a596-014d8a74a2cc · outbound

This paper cites AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models

Reference 90

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.432963Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d94a0715-f578-490c-8a47-ec5e68464ee1 · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Representation Engineering: A Top-Down Approach to AI Transparency

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.438083Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0a5011b-8beb-48f3-b3fd-c6ba8f62a5b6 · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-09T14:47:15.442928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73f8f072-4083-4492-8bce-0cafd19c6284 · outbound

This paper cites Improving alignment and robustness with circuit breakers.

Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Improving alignment and robustness with circuit breakers

Reference 93

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T14:47:35.664850Z

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.

Pith citing papers

Observation 9ac562e3-c5b1-46c0-8e04-7102890168a9 · inbound

Prompt Attacks Reveal Superficial Knowledge Removal in Unlearning Methods cites this paper.

Prompt Attacks Reveal Superficial Knowledge Removal in Unlearning Methods Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-07T04:37:06.683217Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b691c14-d0ac-4408-84b3-1a983d3c9287 · inbound

Anchoring Refusal Direction: Mitigating Safety Risks in Tuning via Projection Constraint cites this paper.

Anchoring Refusal Direction: Mitigating Safety Risks in Tuning via Projection Constraint Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-04T23:10:21.212266Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4017d78-8c14-4e7f-a732-46170f55fb0a · inbound

AntiDote: Bi-level Adversarial Training for Tamper-Resistant LLMs cites this paper.

AntiDote: Bi-level Adversarial Training for Tamper-Resistant LLMs Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-15T16:25:35.442838Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0abc5024-ef5d-4655-95da-9c9b68ade291 · inbound

Downgrade to Upgrade: Optimizer Simplification Enhances Robustness in LLM Unlearning cites this paper.

Downgrade to Upgrade: Optimizer Simplification Enhances Robustness in LLM Unlearning Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:46:17.058220Z

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 9195ab00-e824-41a4-9ef9-dd9eb8448c61 · inbound

RippleBench: Capturing Ripple Effects Using Existing Knowledge Repositories cites this paper.

RippleBench: Capturing Ripple Effects Using Existing Knowledge Repositories Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-03T18:41:33.209039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5ba3b8fb-a6c3-43a7-8dbe-1d7aa8abb55a · inbound

Video Deepfake Abuse: How Company Choices Predictably Shape Misuse Patterns cites this paper.

Video Deepfake Abuse: How Company Choices Predictably Shape Misuse Patterns Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T19:59:01.358929Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 735466bb-16e9-47f0-ad25-d74394d5c890 · inbound

Operationalising the Superficial Alignment Hypothesis via Task Complexity cites this paper.

Operationalising the Superficial Alignment Hypothesis via Task Complexity Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-02T22:49:13.999922Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f63ce60-b49b-432b-94be-3091cd96bea2 · inbound

An Independent Safety Evaluation of Kimi K2.5 cites this paper.

An Independent Safety Evaluation of Kimi K2.5 Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 91

Resolution: verified exact; arxiv_id, observed 2026-05-13T19:43:11.577719Z

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 ec539424-962b-4ec3-b385-e3a4912b5b1d · inbound

Is your algorithm unlearning or untraining? cites this paper.

Is your algorithm unlearning or untraining? Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-11T05:30:59.104808Z

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.