Paper Citation Record · LEDGER

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks

As of 9 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 0 inbound Pith citation observations for arXiv:2607.23015.

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

pith.paper-citation-record.v1
2607.23015 v1

Coverage vector

measured 31 of 31 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-01T03:57:59.499913Z

measured 31 of 31 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+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

31 of 31 outbound references displayed

verified exact0
verified fuzzy0
unresolved31
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 796cccb9-c854-4cd8-bcf7-b97beb564428 · outbound

This paper cites {TwinBreak}: Jailbreaking {LLM}security alignments based on twin prompts,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks {TwinBreak}: Jailbreaking {LLM}security alignments based on twin prompts,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:55.174461Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0542ac60-b3ad-4695-9223-02353c38621f · outbound

This paper cites Neurostrike: Neuron-level attacks on aligned llms,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Neurostrike: Neuron-level attacks on aligned llms,

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:55.328738Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 83ed6338-5e31-4e91-b44b-98cfaf8cb0c8 · outbound

This paper cites Safeneuron: Neuron-level safety alignment for large language models,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Safeneuron: Neuron-level safety alignment for large language models,

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:55.463492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f10810c-19e5-4cbf-be1b-fcaf1ee41741 · outbound

This paper cites Training language models to follow instructions with human feedback,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Training language models to follow instructions with human feedback,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:55.725939Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 763ab5f8-1725-4eb3-9057-b09882cd4d27 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Constitutional AI: Harmlessness from AI Feedback

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:55.883947Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed44964b-fd73-48ec-9382-ab4d85dfd91a · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Direct preference optimization: Your language model is secretly a reward model,

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.021659Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2aad0b3-f51e-4ade-82c0-3d31167e7837 · outbound

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

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.099899Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d757a4c6-6f55-4f11-a82f-ecdbd1b29937 · outbound

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

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Jailbreaking Black Box Large Language Models in Twenty Queries

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.216801Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e74c8c1-5d5b-4f6b-9a27-599da978a466 · outbound

This paper cites Autodan: Generating stealthy jailbreak prompts on aligned large language models,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Autodan: Generating stealthy jailbreak prompts on aligned large language models,

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.314487Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 001c2a94-53fd-4dac-8ff8-bae64a4ebc47 · outbound

This paper cites Fine-tuning aligned language models compromises safety, even when users do not intend to!.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Fine-tuning aligned language models compromises safety, even when users do not intend to!

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.376108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2aae0de6-c913-4bdc-bca5-5a1b869515cd · outbound

This paper cites Shortgpt: Layers in large language models are more redundant than you expect,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Shortgpt: Layers in large language models are more redundant than you expect,

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.418135Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 712b6511-5db3-410d-8381-a8908ea86194 · outbound

This paper cites Blockpruner: Fine-grained pruning for large language models,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Blockpruner: Fine-grained pruning for large language models,

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.556177Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d755635-e9c2-414d-94b3-92b174d6c9ea · outbound

This paper cites Finding safety neurons in large language models,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Finding safety neurons in large language models,

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.718260Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7ba6b8d-cf8f-4b51-bde4-6bf2cda23d72 · outbound

This paper cites NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:56.952079Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c1e420db-5126-429d-bbdc-2a73b18457bd · outbound

This paper cites NeuRel-Attack: Neuron Relearning for Safety Disalignment in Large Language Models.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks NeuRel-Attack: Neuron Relearning for Safety Disalignment in Large Language Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.052252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8d68b254-17c6-462e-8a8a-36d6ab56ad30 · outbound

This paper cites Fine- grained safety neurons with training-free continual projection to reduce llm fine tuning risks,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Fine- grained safety neurons with training-free continual projection to reduce llm fine tuning risks,

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.136280Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ab34325-0e0d-46ee-8b72-53f9ccbb073a · outbound

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

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Refusal in Language Models Is Mediated by a Single Direction

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.180833Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1cc20ce3-f571-4e38-baf4-da20d7705a6c · outbound

This paper cites Beyond surface alignment: Rebuilding llms safety mechanism via probabilistically ablating refusal direction,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Beyond surface alignment: Rebuilding llms safety mechanism via probabilistically ablating refusal direction,

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.344949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49d682ec-adb4-400a-8506-dad91974dee9 · outbound

This paper cites Homotopic language reorganization in the right hemisphere after early left hemisphere injury,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Homotopic language reorganization in the right hemisphere after early left hemisphere injury,

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.470050Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4f09300d-11b9-4e25-b073-12ccf9f44107 · outbound

This paper cites Speak- ing with a single cerebral hemisphere: fmri language organization after hemispherectomy in childhood,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Speak- ing with a single cerebral hemisphere: fmri language organization after hemispherectomy in childhood,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.605873Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e2a5483-bada-415e-b776-1842bf376991 · outbound

This paper cites Qwen2.5 Technical Report.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Qwen2.5 Technical Report

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.763420Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f8c18103-c801-421f-907d-cd234a279709 · outbound

This paper cites The Llama 3 Herd of Models.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks The Llama 3 Herd of Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:57.999000Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75ede269-a074-48c7-ac30-923e19625928 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.210781Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c869549f-6e94-440a-98ca-4b640c0689ff · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Gemma: Open Models Based on Gemini Research and Technology

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.396900Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9530001-e977-441a-8566-efbdeee74f12 · outbound

This paper cites Phi-4 Technical Report.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Phi-4 Technical Report

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.568949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fdd8d583-5c33-46ba-bdc9-e3042292e405 · outbound

This paper cites A strongreject for empty jailbreaks,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks A strongreject for empty jailbreaks,

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.664584Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc742037-5b64-4fba-8eab-13828deb9e35 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Training Verifiers to Solve Math Word Problems

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.789019Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d92abd44-59b5-4ed1-8ab5-9f98cb132053 · outbound

This paper cites HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:58.902932Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97fa6519-e9d4-45ee-be84-5a35048870f5 · outbound

This paper cites Jailbreakbench: An open robustness benchmark for jailbreaking large language models,.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Jailbreakbench: An open robustness benchmark for jailbreaking large language models,

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:59.143074Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29940718-1331-43fe-b547-cc1e9e3f0ff3 · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:59.326603Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation efebfdfd-f71f-4ec9-9b91-c0c93cb989e4 · outbound

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

Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks Truthfulqa: Measuring how models mimic human falsehoods,

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-01T03:57:59.499913Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.