Paper Citation Record · LEDGER

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents

As of 9 August 2026, this Paper Citation Record lists 25 of 25 outbound references and 0 inbound Pith citation observations for arXiv:2608.03844.

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

pith.paper-citation-record.v1
2608.03844 v1

Coverage vector

measured 25 of 25 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-05T11:11:41.019247Z

measured 25 of 25 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+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

25 of 25 outbound references displayed

verified exact3
verified fuzzy6
unresolved16
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation e768d6ae-457f-407f-b277-df9492b5335e · outbound

This paper cites denotes mean cosine similarity to the nearest victim query.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents denotes mean cosine similarity to the nearest victim query

Reference 1

Resolution: verified exact; raw_fallback, observed 2026-08-05T11:11:41.310673Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 80340ae7-0d2e-4be6-ae31-a68efa44ba1c · outbound

This paper cites Evaluating Large Language Models Trained on Code.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Evaluating Large Language Models Trained on Code

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.899044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0eabf312-fd05-4a17-a66d-9f892ec2afd1 · outbound

This paper cites Table 6: Retriever-paradigm generalization on MIMIC- III.(a)Macro RIR@ 4 across nine pairs.(b)ASR under the symmetricall-MiniLM-L6-v2retriever.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Table 6: Retriever-paradigm generalization on MIMIC- III.(a)Macro RIR@ 4 across nine pairs.(b)ASR under the symmetricall-MiniLM-L6-v2retriever

Reference 3

Resolution: verified exact; raw_fallback, observed 2026-08-05T11:11:41.404484Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation fb4edba5-868e-4dbf-9305-7af847e5ac97 · outbound

This paper cites In Findings of the Association for Computational Linguistics: ACL 2025, pages 19796–19821.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents In Findings of the Association for Computational Linguistics: ACL 2025, pages 19796–19821

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:42.082800Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 5f271289-b3ae-430b-a993-cf1c58fb51b5 · outbound

This paper cites Memory in the Age of AI Agents.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Memory in the Age of AI Agents

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.919347Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f0f457d-c628-4c28-8854-d9a825aa991c · outbound

This paper cites Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.924507Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 33bf24a4-063d-4cf9-9c7f-6d62c2996eb2 · outbound

This paper cites arXiv preprint arXiv:2501.18492.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents arXiv preprint arXiv:2501.18492

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.940444Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88c2e2f8-c8e8-4f19-8a87-23e39c6d2a30 · outbound

This paper cites Context manipulation attacks : Web agents are susceptible to corrupted memory.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Context manipulation attacks : Web agents are susceptible to corrupted memory

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.945299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb887256-0e1f-4e48-ae9a-be16a8fbd0db · outbound

This paper cites arXiv preprint arXiv:2602.15344.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents arXiv preprint arXiv:2602.15344

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.954262Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b575b6c-f55f-4b33-9f9e-431bff6775c5 · outbound

This paper cites In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 22315–22339.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 22315–22339

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:42.061190Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 578e49be-e9bd-4f82-b0f5-e7f06ca1deec · outbound

This paper cites arXiv preprint arXiv:2512.16962.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents arXiv preprint arXiv:2512.16962

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.964776Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28778efe-44d3-4554-828f-6148c8da33b0 · outbound

This paper cites Voyager: An Open-Ended Embodied Agent with Large Language Models.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Voyager: An Open-Ended Embodied Agent with Large Language Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.970057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc9973b4-a778-4544-b1ca-7774b7333816 · outbound

This paper cites In International Conference on Learning Representations, volume 2024, pages 32593–32627.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents In International Conference on Learning Representations, volume 2024, pages 32593–32627

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:42.041112Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 97efe2d9-462b-49ee-b3f1-fe9b5b2aa18a · outbound

This paper cites arXiv preprint arXiv:2510.02373.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents arXiv preprint arXiv:2510.02373

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.979534Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21592f37-54c4-4ade-9893-4fd76f3a8811 · outbound

This paper cites In Findings of the Association for Computational Linguistics: ACL 2024, pages 3053–3077.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents In Findings of the Association for Computational Linguistics: ACL 2024, pages 3053–3077

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:42.022862Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation d282220b-6a0b-425e-bd83-569623ba813f · outbound

This paper cites In International Conference on Learning Representations, volume 2024, pages 19036–19066.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents In International Conference on Learning Representations, volume 2024, pages 19036–19066

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:42.006483Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation bc638171-856c-40cf-8aa5-ed5e8c3e7205 · outbound

This paper cites MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.993520Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d9d5a20-c134-41c9-a58c-ceb931ac53ca · outbound

This paper cites The HF Hub (DataInterpreter) row group targets HuggingFace model-identifier redirects in the agent’s from_pretrained call; individual pair definitions appear in Appendix F.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents The HF Hub (DataInterpreter) row group targets HuggingFace model-identifier redirects in the agent’s from_pretrained call; individual pair definitions appear in Appendix F

Reference 24

Resolution: verified exact; raw_fallback, observed 2026-08-05T11:11:41.216772Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 3822f83c-b525-46db-ab69-4e7a15b9a97d · outbound

This paper cites an unresolved cited work.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-08-05T11:11:41.969466Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 58b1adde-2b64-4a1b-82ca-451b6286fbc3 · outbound

This paper cites Program Synthesis with Large Language Models.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Program Synthesis with Large Language Models

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.893326Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6aa3b3f-9640-4083-a2df-50b87aaba820 · outbound

This paper cites out of stock.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents out of stock

Reference 2022

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T11:11:41.988261Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 38978088-8e05-4ea8-b505-c6a0df712d79 · outbound

This paper cites Retrieval-Augmented Generation for Large Language Models: A Survey.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Retrieval-Augmented Generation for Large Language Models: A Survey

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.909397Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8a49a66-ed81-4302-aab3-906d9caea3e7 · outbound

This paper cites RAP: Retrieval-Augmented Planning with Contextual Memory for Multimodal LLM Agents.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents RAP: Retrieval-Augmented Planning with Contextual Memory for Multimodal LLM Agents

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.930421Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc1df7c6-0c19-454d-b6f0-3c39fbc4c786 · outbound

This paper cites Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.904047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 252e3da5-5bdb-4bae-9db2-6228be2a13d7 · outbound

This paper cites Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems.

MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-05T11:11:40.935774Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.