Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-12T00:37:02.534015Z
Paper Citation Record · LEDGER
As of 22 August 2026, this Paper Citation Record lists 49 of 49 outbound references and 0 inbound Pith citation observations for arXiv:2608.08027.
A citation records a reference. It does not transfer a finding from one paper to another.
Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-12T00:37:02.534015Z
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-22T06:32:14.747728+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links
A source-named dated measurement, never combined with another source.
Source: cited_works
49 of 49 outbound references displayed
External citation measurements
No source-named external measurement is stored.
Observation affd7774-c9fa-4050-9dc8-21b095188d5c · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes A Survey of Large Language Models
Reference 1
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 5bfd5039-8f75-4a61-bf84-6add7e7ebaf7 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Formalizing and benchmarking prompt injection attacks and defenses,
Reference 2
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 1d712126-7fb5-4736-87bb-c28c34b4a3c9 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes OW ASP top 10 for large language model applications,
Reference 3
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 8b7c1f00-ab62-47ff-9097-0d75c5385042 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Delimiters won’t save you from prompt injection,
Reference 4
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 3c495ea6-a1db-492c-a99a-53a67b454344 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Sandwich defense,
Reference 5
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 9faf0654-c4c9-4ff9-bf63-7f2339053561 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Instruction defense,
Reference 6
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 3ef082e4-a440-4fb6-9df3-e375e6f9a1a3 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Prompt-guard-86m,
Reference 7
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 3fd2f2d0-cba3-41bc-9869-7399d53d36bd · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Fine-tuned deberta-v3-base for prompt injection detection,
Reference 8
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 bf0ffde1-fe28-41ea-be07-379142738bdf · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Signed-prompt: A new approach to prevent prompt injection attacks against llm-integrated applications,
Reference 9
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 1bd8d7e8-3662-455a-bb52-40974ff9b358 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Attention tracker: Detecting prompt injection attacks in llms,
Reference 10
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 217f530d-ceed-4699-a3a1-6e4ea54d8d22 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models
Reference 11
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 46f822c2-6291-4a46-9a9a-a5d397edea0d · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Piguard: Prompt injection guardrail via mitigating overdefense for free,
Reference 12
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 690294cf-0e4b-431c-9802-24a9a8caee98 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Securing LLM systems against prompt injec- tion,
Reference 13
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 000d4c6b-4b4e-4344-b710-d51afa8cdad8 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Prompt injection attacks against GPT-3,
Reference 14
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 93c60bfb-083d-4478-a0de-a297b3afb12a · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Universal and Transferable Adversarial Attacks on Aligned Language Models
Reference 15
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 74b273c7-1791-4886-8fb5-4878fa68b026 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Autodan: Generating stealthy jailbreak prompts on aligned large language models,
Reference 16
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 74bd7a97-b3b4-4fb4-b79f-3e78d57a2076 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes x.com — x.com,
Reference 17
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 3e5b6a83-baaa-4a28-925a-086bab7c0e50 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Understanding intermediate layers using linear classifier probes
Reference 18
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 62bdad09-b8ec-4a9b-9c1e-37e5aaf7a2f1 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Probing classifiers: Promises, shortcomings, and ad- vances,
Reference 19
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation c7dabdcc-db6f-47d0-8272-aba8ad2cf329 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes In-context Learning and Induction Heads
Reference 20
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 1ef72592-bcbb-4ed1-a36a-4e91c77063e6 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Retrieval head mecha- nistically explains long-context factuality,
Reference 21
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 08fce25b-1352-446e-92cf-624c2991a22c · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Regularization and variable selection via the elastic net,
Reference 22
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 c86b8ca2-80eb-494d-82be-fc87b786ec82 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Recursive deep models for semantic compositionality over a sentiment treebank,
Reference 23
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 05cc63a6-7251-452d-8fca-643a07d0ac3a · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Foundation
Reference 24
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 28f3d2c0-9db7-4df9-a7ad-c299296739a0 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Don’t give me the details, just the summary! topic-aware convolutional neural networks for extreme summarization,
Reference 25
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 8d817112-e822-4279-9920-a4a37b1dd169 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Squad: 100,000+ questions for machine comprehension of text,
Reference 26
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 a0121950-8fea-4a9e-825c-d2975bc2bf5e · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Ignore Previous Prompt: Attack Techniques For Language Models
Reference 27
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 2f50e9b5-c29f-4cd3-968c-2f403ed13af4 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Llama 2: Open Foundation and Fine-Tuned Chat Models
Reference 28
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 8e5e477e-be3f-422d-ae2c-de826c268565 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Defending Against Indirect Prompt Injection Attacks With Spotlighting
Reference 29
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation ccacb7c5-127f-44a3-b503-0f802982fd12 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Prompt programming for large language models: Beyond the few-shot paradigm,
Reference 30
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 0da2a6ea-a9c5-4e93-92ae-1a6a8c19ed30 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Qwen3 Technical Report
Reference 31
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 95d9d1f7-da07-4fef-9f58-7bd1ed0ceea4 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Team, “Gemma,” 2024
Reference 32
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 8e05a92c-9589-4af0-a618-64eeaa2375fc · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes The Llama 3 Herd of Models
Reference 33
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 67e34488-9820-4136-a5a7-07074b96da0f · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Mistral-nemo-instruct-2407,
Reference 34
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 d53a7dea-10e5-4d0f-861b-45c50f50e580 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study
Reference 35
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 882902e1-2a6d-492b-bfd2-d57635407841 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes ”do anything now
Reference 36
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 497259d6-83fa-4e43-ad64-f17e7987c085 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Ignore this title and hackaprompt: Exposing systemic vulnerabilities of llms through a global prompt hacking competition,
Reference 37
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 7fc2e783-097f-4d9f-8e10-4341c9aefb57 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Scikit-learn: Machine learning in python,
Reference 38
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 116331e2-f4ea-4982-8a05-28b4f2df9200 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes sklearn.linear model.LogisticRegression,
Reference 39
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 12613e14-77dc-487c-aa1f-c2611f90e198 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Transformers: State-of- the-art natural language processing,
Reference 40
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 588eea50-311a-4c81-a896-f928d719086e · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Positive
Reference 43
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 d413e9c4-beeb-41ca-90df-859e16de5204 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Format: Output only the Chinese translation
Reference 44
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 01b5a6bf-1711-4120-9c64-b28f8db1a63a · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Format: Output only a 2-3 sentence summary
Reference 45
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 0bfba195-c095-482a-9610-e992830e177a · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Not mentioned
Reference 46
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 a3a5a0e6-eb62-4e81-9fdf-650f7e987fe2 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes ignore previous instructions
Reference 47
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 e15cf449-8db5-47aa-acab-43bd14c3a54c · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Unresolved cited work
Reference 48
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 7c266ef6-b9dd-4e8c-a515-563ffd8de2b7 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Unresolved cited work
Reference 49
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 7d6db094-71d1-4e9f-890e-29bf69055ccc · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Available: https://simonwillison.net/2023/May/11/ delimiters-wont-save-you/
Reference 2023
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 91bc80c4-86ee-4578-a22a-76ef79b88646 · outbound
BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes Available: https://scikit-learn.org/stable/modules/ generated/sklearn.linear model.LogisticRegression.html
Reference 2026
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.
No inbound Pith citation observations are available.