Paper Citation Record · LEDGER

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering

As of 9 August 2026, this Paper Citation Record lists 28 of 28 outbound references and 4 inbound Pith citation observations for arXiv:2506.06384.

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

pith.paper-citation-record.v1
2506.06384 v1

Coverage vector

measured 28 of 28 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T10:40:02.073583Z

measured 32 of 32 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 4 of 4 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-04T04:34:33.183294Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-06-29T18:43:50.549550Z

Reference resolution

28 of 28 outbound references displayed

verified exact1
verified fuzzy13
unresolved14
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation c4ba760a-c8d5-436b-a4cb-2ce710cb6e8c · outbound

This paper cites Train- ing language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Train- ing language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.947408Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c117b332-c730-4fcd-9a42-205d554d5a50 · outbound

This paper cites Palm: Scaling language modeling with pathways.Journal of Machine Learning Research, 24(240):1–113, 2023.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Palm: Scaling language modeling with pathways.Journal of Machine Learning Research, 24(240):1–113, 2023

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.952724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3a47ffe-2f99-498b-9b08-4b510b03df8c · outbound

This paper cites Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.957240Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4664779d-693b-4677-bab2-09e320795829 · outbound

This paper cites MusicLM: Generating Music From Text.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering MusicLM: Generating Music From Text

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.962885Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ea882a2-982e-4262-b2a4-4fab70870866 · outbound

This paper cites Audiogpt: Under- standing and generating speech, music, sound, and talking head.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Audiogpt: Under- standing and generating speech, music, sound, and talking head

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.488225Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 6d042715-2234-494d-a950-556ea77a2db5 · outbound

This paper cites Understanding large-language model (llm)-powered human-robot interaction.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Understanding large-language model (llm)-powered human-robot interaction

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.474369Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 178c40c7-2183-423b-98ad-2a0b4039ebf5 · outbound

This paper cites Security and privacy challenges of large language models: A survey.ACM Computing Surveys, 57(6):1– 39, 2025.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Security and privacy challenges of large language models: A survey.ACM Computing Surveys, 57(6):1– 39, 2025

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.458542Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5240f838-8660-41d9-887e-5c30325b5db7 · outbound

This paper cites A survey on large language model (llm) security and privacy: The good, the bad, and the ugly.High-Confidence Computing, page 100211, 2024.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering A survey on large language model (llm) security and privacy: The good, the bad, and the ugly.High-Confidence Computing, page 100211, 2024

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.981949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 068b22bf-1c41-4592-8d10-4e81798844b7 · outbound

This paper cites Owasp top 10 list for large language models, 2024.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Owasp top 10 list for large language models, 2024

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.433856Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 4b39447d-c473-4624-8fac-011f4335ed64 · outbound

This paper cites Ignore previous prompt: Attack techniques for lan- guage models.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Ignore previous prompt: Attack techniques for lan- guage models

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.416679Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 4ac2a4fc-ee9f-4d76-9172-1b61d398450b · outbound

This paper cites Not what you’ve signed up for: Compromising real-world llm-integrated applications with indirect prompt injection.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Not what you’ve signed up for: Compromising real-world llm-integrated applications with indirect prompt injection

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:01.995703Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 714dd145-fbfc-4f49-826f-39b2b5836cc7 · outbound

This paper cites Formalizing and benchmarking prompt injection attacks and defenses.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Formalizing and benchmarking prompt injection attacks and defenses

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.000141Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86d29bf5-8343-41e5-b7b6-6b65800b7715 · outbound

This paper cites Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-08-07T10:40:02.196160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation c35267ab-5f2f-440d-bad6-77b7dcd9db51 · outbound

This paper cites Many-shot jail- breaking.Advances in Neural Information Processing Systems, 37:129696–129742, 2025.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Many-shot jail- breaking.Advances in Neural Information Processing Systems, 37:129696–129742, 2025

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.382294Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5c5cbb62-22d0-4aba-8e65-fea1c7bffd8c · outbound

This paper cites Deep reinforcement learning from human preferences.Advances in neural information processing systems, 30, 2017.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Deep reinforcement learning from human preferences.Advances in neural information processing systems, 30, 2017

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.014098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1da5feb4-3636-48f3-875d-a58907861f90 · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model.Advances in Neural Information Processing Systems, 36:53728–53741, 2023.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Direct preference optimization: Your language model is secretly a reward model.Advances in Neural Information Processing Systems, 36:53728–53741, 2023

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.019307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 15018687-e03a-4e0c-87d9-376046691d05 · outbound

This paper cites GPT-4 Technical Report.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering GPT-4 Technical Report

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.023821Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be004a1b-f12e-4353-8d4c-f677a2349ff8 · outbound

This paper cites DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.028913Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a797b61-fe85-4db6-ba0a-7a458322ca1b · outbound

This paper cites fmops/distilbert-prompt-injection, 2024.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering fmops/distilbert-prompt-injection, 2024

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.346150Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation d6e28dd2-6c8c-45ec-9235-8c671ff01b87 · outbound

This paper cites Fine-tuned deberta-v3-base for prompt injection detection, 2024.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Fine-tuned deberta-v3-base for prompt injection detection, 2024

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.330233Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation c8af2019-9508-4de7-8345-8407d8fbd83e · outbound

This paper cites Safeguard: A benchmark suite for evaluating attacks and defenses on llm safety, 2023.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Safeguard: A benchmark suite for evaluating attacks and defenses on llm safety, 2023

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.311868Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 347b9855-fd2b-4166-9014-f50b5208fba5 · outbound

This paper cites InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.046929Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3f048285-c3ef-412a-99b6-0e3ac8b8e7ea · outbound

This paper cites Apply- ing pre-trained multilingual bert in embeddings for improved malicious prompt injection attacks detection.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Apply- ing pre-trained multilingual bert in embeddings for improved malicious prompt injection attacks detection

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.295665Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 69f768fe-609f-4c44-ac91-940c42493a2c · outbound

This paper cites StruQ: Defending Against Prompt Injection with Structured Queries.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering StruQ: Defending Against Prompt Injection with Structured Queries

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.055581Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e4b89cd-86d9-4c36-bfd5-8632b3b46801 · outbound

This paper cites Jatmo: Prompt injection defense by task- specific finetuning.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Jatmo: Prompt injection defense by task- specific finetuning

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.279552Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation ed4ee657-a579-4f77-ab4e-842425669b6d · outbound

This paper cites Llm self defense: By self examina- tion, llms know they are being tricked.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Llm self defense: By self examina- tion, llms know they are being tricked

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.263645Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f13cddc9-c7e3-40ba-b56c-d5d270f17f3d · outbound

This paper cites Defending chatgpt against jailbreak attack via self- reminders.Nature Machine Intelligence, 5(12):1486–1496, 2023.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering Defending chatgpt against jailbreak attack via self- reminders.Nature Machine Intelligence, 5(12):1486–1496, 2023

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T10:40:02.248336Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f2d2bbe2-ea9e-4431-9c65-a349c101150b · outbound

This paper cites CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models.

Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T10:40:02.073583Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 2f6a1569-d3d0-47f7-b50b-a4eacfd6957d · inbound

Indirect Prompt Injection in the Wild: An Empirical Study of Prevalence, Techniques, and Objectives cites this paper.

Indirect Prompt Injection in the Wild: An Empirical Study of Prevalence, Techniques, and Objectives Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering

Reference 94

Resolution: verified exact; arxiv_id, observed 2026-05-12T09:51:29.356507Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 0792b659-996d-4496-8912-a8fd2971fa16 · inbound

Prompt Injection Detection is Regime-Dependent: A Deployment-Aware Evaluation with Interpretable Structural Signals cites this paper.

Prompt Injection Detection is Regime-Dependent: A Deployment-Aware Evaluation with Interpretable Structural Signals Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-06-29T18:43:50.550887Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f6fdcd99-579b-4d33-9c1c-929a97a216a0 · inbound

CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs cites this paper.

CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T08:55:49.093096Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4d5d6f6-8c08-4815-9c41-571f3634349c · inbound

CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs cites this paper.

CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs Detection Method for Prompt Injection by Integrating Pre-trained Model and Heuristic Feature Engineering

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-04T04:34:33.183294Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.