Paper Citation Record · LEDGER

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs

As of 24 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 4 inbound Pith citation observations for arXiv:2501.18626.

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

pith.paper-citation-record.v1
2501.18626 v4

Coverage vector

measured 31 of 31 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-10T13:53:10.999594Z

measured 35 of 35 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-23T06:30:58.430688+00:00

measured 4 of 4 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-15T17:12:27.776350Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-22T23:32:15.547339Z

Reference resolution

31 of 31 outbound references displayed

verified exact1
verified fuzzy0
unresolved30
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 3b069cd7-e081-4bec-ac55-eebcb87a4939 · outbound

This paper cites Does Refusal Training in LLMs Generalize to the Past Tense?.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Does Refusal Training in LLMs Generalize to the Past Tense?

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.916955Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23002035-bef3-4797-a923-04f22aee1eba · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.921490Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 042f2853-3088-45a8-aaac-7252d334bd5c · outbound

This paper cites JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.923889Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 882387c2-eda5-4729-8a8c-91dfb740634a · outbound

This paper cites Safe RLHF: Safe Reinforcement Learning from Human Feedback.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Safe RLHF: Safe Reinforcement Learning from Human Feedback

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.927154Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1bd78ad-d57e-4e45-a751-bbc4dda800c5 · outbound

This paper cites Building Guardrails for Large Language Models.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Building Guardrails for Large Language Models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.930539Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c8989374-b901-4edf-9035-23c83495dff9 · outbound

This paper cites Safeguarding Large Language Models: A Survey.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Safeguarding Large Language Models: A Survey

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.933726Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc78a156-1abb-4934-9031-be723e29ce18 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 7

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.427583Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 647a6428-80be-4819-9eba-3a2f173ecdf9 · outbound

This paper cites Gemma 2: Improving Open Language Models at a Practical Size.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Gemma 2: Improving Open Language Models at a Practical Size

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.939333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f48098f6-712a-4a75-a291-d2f115c90b7f · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 9

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.420951Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 2a19c080-0f53-46d8-8cf9-fca5c74bcbba · outbound

This paper cites BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.944222Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 597a9711-fba7-4108-862f-9f757296b2f3 · outbound

This paper cites Single Character Perturbations Break LLM Alignment.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Single Character Perturbations Break LLM Alignment

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-08-10T13:53:11.241158Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 07d0002d-e631-41e8-9474-aedc25cb55ba · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 12

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.413947Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation db1a50b7-41b6-444a-b628-0b4e09d9dd28 · outbound

This paper cites Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.952386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 414db0aa-2e9b-45ac-8cac-5ad4a6b01815 · outbound

This paper cites Tree of Attacks: Jailbreaking Black-Box LLMs Automatically.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Tree of Attacks: Jailbreaking Black-Box LLMs Automatically

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.955205Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2843845-7d5c-4311-9773-a9786f78c2f6 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.407634Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 5e7e662b-aebf-460c-8456-7850ac22fd68 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.400980Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation edb41915-0583-433b-8db2-4aaf4f1f779d · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 17

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.394447Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 4550525d-c4cf-4bd7-b094-71c336fe6311 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 18

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.386864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 1f5f8411-04e3-4d97-a8da-aa007534dd25 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.380330Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation d996eb7a-c653-4074-8f10-2eb43f381f03 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 20

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.372596Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation bd221061-12f2-4cfa-978a-c44a7c9e5a80 · outbound

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

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.973355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39032472-e589-4a3a-87ad-4d434a19a32b · outbound

This paper cites do anything now.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs do anything now

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.975928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 554f58e4-b208-4242-b7ae-3201d02cd7b0 · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.365515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 01e275ba-f362-444f-95ad-03271447018e · outbound

This paper cites The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.980458Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02c7a1fa-7a82-42ee-ab41-5dd45ea8bcab · outbound

This paper cites CYBERSECEVAL 3: Advancing the Evaluation of Cybersecurity Risks and Capabilities in Large Language Models.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs CYBERSECEVAL 3: Advancing the Evaluation of Cybersecurity Risks and Capabilities in Large Language Models

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.983055Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75fa16ee-f3bf-4d7b-8c33-0918fe9d9b6a · outbound

This paper cites Jailbroken: How Does LLM Safety Training Fail?.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Jailbroken: How Does LLM Safety Training Fail?

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.985478Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e865d0b-8f1c-458d-8295-91538a363dbb · outbound

This paper cites an unresolved cited work.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-08-10T13:53:11.358302Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation d93a15bb-7882-4f30-bd4f-ef0b857eaebe · outbound

This paper cites An LLM can Fool Itself: A Prompt-Based Adversarial Attack.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs An LLM can Fool Itself: A Prompt-Based Adversarial Attack

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.991708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 110524e7-28a8-4eb9-a44d-05c89944aca7 · outbound

This paper cites Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.994325Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a8d6227-1051-4734-8e63-64b05a5429fa · outbound

This paper cites online" 'onlinestring :=.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs online" 'onlinestring :=

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.996810Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e52257d-8ebf-4cb3-a449-7f97d3fdcd2a · outbound

This paper cites write newline.

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs write newline

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-10T13:53:10.999594Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 642b46d7-505c-43bd-b4f6-d10351af91ce · inbound

Toxicity Detection Should Measure Contextual Harm, Not Text-Intrinsic Badness cites this paper.

Toxicity Detection Should Measure Contextual Harm, Not Text-Intrinsic Badness The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-05-22T23:32:15.549899Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.

Observation 2661a4dc-7c8d-4897-a17c-5d5a869f2b21 · inbound

TRIDENT: Benchmarking LLM Safety in Finance, Medicine, and Law cites this paper.

TRIDENT: Benchmarking LLM Safety in Finance, Medicine, and Law The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-06T15:04:21.357639Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6a1439e-624a-4b3c-80d5-9c78bf3d68ed · inbound

POT: Inducing Overthinking in LLMs via Black-Box Iterative Optimization cites this paper.

POT: Inducing Overthinking in LLMs via Black-Box Iterative Optimization The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-15T17:12:27.776350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad93d65a-7726-4ec3-9e37-9d3dc4335df1 · inbound

LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems cites this paper.

LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-04T17:46:12.415475Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.