Paper Citation Record · LEDGER

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders

As of 22 August 2026, this Paper Citation Record lists 35 of 35 outbound references and 0 inbound Pith citation observations for arXiv:2605.20759.

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

pith.paper-citation-record.v1
2605.20759 v1

Coverage vector

measured 35 of 35 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-21T04:33:04.629491Z

measured 35 of 35 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 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

35 of 35 outbound references displayed

verified exact18
verified fuzzy17
unresolved0
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 7a28163e-da0a-469f-9537-64dd70fdb94f · outbound

This paper cites Language Models are Few-Shot Learners.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Language Models are Few-Shot Learners

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.738833Z

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 9c6ad8c0-cfd5-4c01-b552-8b22adafdde6 · outbound

This paper cites Training Language Models to Follow Instructions with Human Feedback.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Training Language Models to Follow Instructions with Human Feedback

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.763965Z

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 807fc202-5b67-4657-acc5-3988c1560a2f · outbound

This paper cites Graph Neural Networks for Financial Fraud Detection: A Review.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Graph Neural Networks for Financial Fraud Detection: A Review

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.769125Z

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 d9d84d10-5be7-4d1e-8223-d2a5948603f1 · outbound

This paper cites A Comprehensive Survey on Graph Neural Networks.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders A Comprehensive Survey on Graph Neural Networks

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.755904Z

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 3c4ee06d-924e-49ae-866a-7de8701c096a · outbound

This paper cites Survey on Graph Neural Networks.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Survey on Graph Neural Networks

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.758433Z

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 2f61506a-f6fa-4b51-9121-84d0b11afd8c · outbound

This paper cites Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.766791Z

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 13010008-c3bb-4ddc-9719-01fefc8641d3 · outbound

This paper cites A Temporal Graph Network Algorithm for Detecting Fraudulent Transactions on Online Payment Platforms.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders A Temporal Graph Network Algorithm for Detecting Fraudulent Transactions on Online Payment Platforms

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.771816Z

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 d55e98c5-8749-4815-b7ae-672d89bbe8d5 · outbound

This paper cites Autonomous and Teleoperation Control of a Drawing Robot Avatar.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Autonomous and Teleoperation Control of a Drawing Robot Avatar

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.799394Z

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 98f30947-2a65-47b8-a9a4-0b9a66d7012c · outbound

This paper cites CaT-GNN: Enhancing Credit Card Fraud Detection via Causal Temporal Graph Neural Networks.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders CaT-GNN: Enhancing Credit Card Fraud Detection via Causal Temporal Graph Neural Networks

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.792692Z

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 6e75378b-366e-4731-a79d-173209fe50d8 · outbound

This paper cites Temporal Graph Networks for Deep Learning on Dynamic Graphs.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Temporal Graph Networks for Deep Learning on Dynamic Graphs

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.789960Z

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 c9187cc1-967b-459b-8075-0a0d79a7aa6d · outbound

This paper cites Inductive Representation Learning on Temporal Graphs.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Inductive Representation Learning on Temporal Graphs

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.749440Z

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 da766003-663c-4413-8e78-c9b330a356a3 · outbound

This paper cites DetoxBench: Benchmarking Large Language Models for Multitask Fraud & Abuse Detection.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders DetoxBench: Benchmarking Large Language Models for Multitask Fraud & Abuse Detection

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.787066Z

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 539d436b-7d58-49f5-899e-366689babc30 · outbound

This paper cites Fraud-R1 : A Multi-Round Benchmark for Assessing the Robustness of LLM Against Augmented Fraud and Phishing Inducements.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Fraud-R1 : A Multi-Round Benchmark for Assessing the Robustness of LLM Against Augmented Fraud and Phishing Inducements

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.795729Z

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 f44e2550-66fe-4be8-bce1-a6ddf263678f · outbound

This paper cites Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.779596Z

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 f8806542-5da1-4641-a848-2517a42c04e0 · outbound

This paper cites Large Language Model Safety: A Holistic Survey.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Large Language Model Safety: A Holistic Survey

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.782996Z

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 b294d46f-e5ef-4149-ad16-a1d652802e67 · outbound

This paper cites A Survey on Large Language Model Security and Privacy: The Good, The Bad, and The Ugly.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders A Survey on Large Language Model Security and Privacy: The Good, The Bad, and The Ugly

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.746619Z

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 d9123fc9-56e1-4964-9b6c-46b576518115 · outbound

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

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Jailbroken: How Does LLM Safety Training Fail?

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.728419Z

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 8266d6b1-03b8-4959-8a40-0a55727f0cdd · outbound

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

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.776948Z

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 b2c180c7-fa32-4275-a623-82e6c9b64f93 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Constitutional AI: Harmlessness from AI Feedback

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.749680Z

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 076c2166-9edf-472b-8558-e4f884b7c410 · outbound

This paper cites TrustLLM: Trustworthiness in Large Language Models.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders TrustLLM: Trustworthiness in Large Language Models

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.741561Z

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 10ddc767-05bc-4979-a342-22609dde0641 · outbound

This paper cites DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.743993Z

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 1764cc2f-689d-4641-bc30-6bc5fbea9765 · outbound

This paper cites Ignore Previous Prompt: Attack Techniques For Language Models.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Ignore Previous Prompt: Attack Techniques For Language Models

Reference 22

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.755187Z

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 7d75c099-f504-4b0c-b867-ef256ec44bb8 · outbound

This paper cites Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection

Reference 23

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.757956Z

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 c8f8363d-7ffd-495f-ba5d-fd5cfed5e44a · outbound

This paper cites Semi-Supervised Classification with Graph Convolutional Networks.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Semi-Supervised Classification with Graph Convolutional Networks

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.761334Z

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 1692f62e-cbc7-4b34-a469-62b47a473e68 · outbound

This paper cites Graph Attention Networks.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Graph Attention Networks

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.733458Z

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 a23dad03-0af0-49d7-b3a2-5347e2602ad4 · outbound

This paper cites Inductive Representation Learning on Large Graphs.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Inductive Representation Learning on Large Graphs

Reference 26

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.752504Z

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 daf4da90-bec8-4ec6-b4f0-bbaaa84c90be · outbound

This paper cites Graph Learning: A Survey.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Graph Learning: A Survey

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.725671Z

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 02f9aaeb-7e7e-4f74-8546-f62f739c9caa · outbound

This paper cites Dual-mode rf cavity: design, tuning and performance.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Dual-mode rf cavity: design, tuning and performance

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.761447Z

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 9497bb7a-0405-4317-991f-d6ffb3c05474 · outbound

This paper cites A Survey on Augmenting Knowledge Graphs with Large Language Models: Methods, Challenges, and Future Directions.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders A Survey on Augmenting Knowledge Graphs with Large Language Models: Methods, Challenges, and Future Directions

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.723162Z

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 fb1e9f18-d0d7-4f82-862f-1db625769434 · outbound

This paper cites LLM-Based Multi-Hop Question Answering with Knowledge Graph Integration in Evolving Environments.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders LLM-Based Multi-Hop Question Answering with Knowledge Graph Integration in Evolving Environments

Reference 30

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.764502Z

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 8110edc2-6e04-4029-83e7-c5bd740cb521 · outbound

This paper cites Beyond Knowledge to Agency: Evaluating Expertise, Autonomy, and Integrity in Finance with CNFinBench.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Beyond Knowledge to Agency: Evaluating Expertise, Autonomy, and Integrity in Finance with CNFinBench

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-06-01T02:02:22.916902Z

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 143a970c-f0a5-4aa4-ba43-6b9bd7314365 · outbound

This paper cites SafeDialBench: A Fine-Grained Safety Benchmark for Large Language Models in Multi-Turn Dialogues.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders SafeDialBench: A Fine-Grained Safety Benchmark for Large Language Models in Multi-Turn Dialogues

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-05-21T04:33:57.767962Z

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 c955fa02-e364-456a-88e0-fda6e9f0c36f · outbound

This paper cites FinSafetyBench: Evaluating LLM Safety in Real-World Financial Scenarios.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders FinSafetyBench: Evaluating LLM Safety in Real-World Financial Scenarios

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.771369Z

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 6bc4d6c8-8213-40f3-9ebb-06602071f958 · outbound

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

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-05-21T04:33:57.773747Z

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 da1bc521-9318-4882-8057-5ae13589bb7d · outbound

This paper cites Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena.

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T04:34:34.736075Z

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

No inbound Pith citation observations are available.