Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure | Paper Citation Record

Observation d08fbfda-b34e-4243-8c83-f0030f19dcda · outbound

This paper cites Get my drift? catching llm task drift with activation deltas.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Get my drift? catching llm task drift with activation deltas

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.266213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66fab6d8-702a-4ecc-838a-c48f2fc9e688 · outbound

This paper cites Prompt leakage effect and mitigation strategies for multi-turn llm applications.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Prompt leakage effect and mitigation strategies for multi-turn llm applications

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.268994Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91de3c53-35a6-415b-95ed-0cf40ab86c8f · outbound

This paper cites Understanding intermediate layers using linear classifier probes.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Understanding intermediate layers using linear classifier probes

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.271636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff90bb45-4397-4ff9-956f-2edd8c5bd96a · outbound

This paper cites IPIGuard : A novel tool dependency graph-based defense against indirect prompt injection in LLM agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure IPIGuard : A novel tool dependency graph-based defense against indirect prompt injection in LLM agents

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.274153Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2e9699b-5c2f-453f-b38f-89a9b3e2222b · outbound

This paper cites Jailbreaking leading safety-aligned LLM s with simple adaptive attacks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Jailbreaking leading safety-aligned LLM s with simple adaptive attacks

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.276562Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8cc9763a-8480-4557-8eb1-bb1e64a55016 · outbound

This paper cites Many-shot jailbreaking.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Many-shot jailbreaking

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.279035Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30cb68a9-404a-4d88-919d-0e074d8523f5 · outbound

This paper cites Refusal in language models is mediated by a single direction.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Refusal in language models is mediated by a single direction

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.281666Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1dad014f-f556-4ac1-b8e3-dad75af88d13 · outbound

This paper cites Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.283796Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e0e27b3-5bb3-4cae-92ed-2fa41f461abf · outbound

This paper cites Probing classifiers: Promises, shortcomings, and advances.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Probing classifiers: Promises, shortcomings, and advances

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.287306Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e8ce96e-6279-4039-9eed-96ffd287cead · outbound

This paper cites Language models can explain neurons in language models.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Language models can explain neurons in language models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.289469Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 63e5737b-96ba-42cb-82fd-4e0fb201b452 · outbound

This paper cites Bogdan, Uzay Macar, Neel Nanda, and Arthur Conmy.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Bogdan, Uzay Macar, Neel Nanda, and Arthur Conmy

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.291652Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6a263c8-2f67-4aae-9e20-a3d706d5937e · outbound

This paper cites Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.293790Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a28847c1-2e00-4907-a14e-ef00c900d453 · outbound

This paper cites Towards monosemanticity: Decomposing language models with dictionary learning.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Towards monosemanticity: Decomposing language models with dictionary learning

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.296237Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 63baef44-47d6-4ed6-83ab-0f7c47353381 · outbound

This paper cites Discovering latent knowledge in language models without supervision.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Discovering latent knowledge in language models without supervision

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.298288Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56473867-639c-4524-b872-3b520e6d560d · outbound

This paper cites Vpi-bench: Visual prompt injection attacks for computer-use agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Vpi-bench: Visual prompt injection attacks for computer-use agents

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.300455Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 72eeef71-88fe-4937-8ebc-8b431c533a5d · outbound

This paper cites Ghostei-bench: Do mobile agents resilience to environmental injection in dynamic on-device environments? In ICLR, 2026.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Ghostei-bench: Do mobile agents resilience to environmental injection in dynamic on-device environments? In ICLR, 2026

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.302659Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4008ff70-ac65-45f8-949c-28426be73e33 · outbound

This paper cites Struq: Defending against prompt injection with structured queries.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Struq: Defending against prompt injection with structured queries

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.662574Z

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 89feb34e-ed98-4c2f-be38-e21986457ec7 · outbound

This paper cites Secalign: Defending against prompt injection with preference optimization.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Secalign: Defending against prompt injection with preference optimization

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.307016Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a36b39f7-071b-415a-89e2-6df75d2f875a · outbound

This paper cites HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimization.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimization

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.309027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 15ec6143-48ab-4c62-a8cc-746c2cf77db2 · outbound

This paper cites LlamaFirewall: An open source guardrail system for building secure AI agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure LlamaFirewall: An open source guardrail system for building secure AI agents

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.311523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b234a609-1001-450b-9c75-f1abcba9a8f2 · outbound

This paper cites Constitutional classifiers++: Efficient production-grade defenses against universal jailbreaks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Constitutional classifiers++: Efficient production-grade defenses against universal jailbreaks

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.313974Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0c0b495-3417-41f4-986a-d71828e8602e · outbound

This paper cites Agentdojo: A dynamic environment to evaluate prompt injection attacks and defenses for LLM agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Agentdojo: A dynamic environment to evaluate prompt injection attacks and defenses for LLM agents

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.650024Z

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 8ac70048-ae95-416f-a63d-231248544b44 · outbound

This paper cites Defeating Prompt Injections by Design.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Defeating Prompt Injections by Design

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.318137Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 966911ef-668d-4ccc-ab43-9189468bdaca · outbound

This paper cites `` I ' ve decided to leak'': Probing internals behind prompt leakage intents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure `` I ' ve decided to leak'': Probing internals behind prompt leakage intents

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.642090Z

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 d0d96b01-9b34-4a1c-b49f-a8102a1655d3 · outbound

This paper cites Safesearch: Automated red-teaming of LLM -based search agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Safesearch: Automated red-teaming of LLM -based search agents

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.633456Z

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 6d6264ae-f349-4461-8f4e-b0b2d5f9cefa · outbound

This paper cites Memory injection attacks on llm agents via query-only interaction.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Memory injection attacks on llm agents via query-only interaction

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.626838Z

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 2eaef61b-b7c2-4df6-be65-02cc20df039d · outbound

This paper cites Transcoders find interpretable llm feature circuits.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Transcoders find interpretable llm feature circuits

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.619747Z

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 0b403446-6ee0-4ab2-aecb-e7567a98b1cf · outbound

This paper cites How vulnerable are ai agents to indirect prompt injections? insights from a large-scale public competition.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure How vulnerable are ai agents to indirect prompt injections? insights from a large-scale public competition

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.329002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0cbdf7b-546c-42aa-b51a-009e357d52dc · outbound

This paper cites WASP : Benchmarking web agent security against prompt injection attacks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure WASP : Benchmarking web agent security against prompt injection attacks

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.612970Z

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 64116890-5e9e-4b32-90d6-122cb6145892 · outbound

This paper cites Patchscopes: A unifying framework for inspecting hidden representations of language models.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Patchscopes: A unifying framework for inspecting hidden representations of language models

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.605163Z

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 19713e46-219d-41af-8531-70043a946e9f · outbound

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

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Not what you've signed up for: Compromising real-world llm-integrated applications with indirect prompt injection

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.597790Z

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 52dae4b0-414e-4e78-8dc9-83e6b29ee94d · outbound

This paper cites Agent smith: a single image can jailbreak one million multimodal llm agents exponentially fast.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Agent smith: a single image can jailbreak one million multimodal llm agents exponentially fast

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.590704Z

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 151fd113-bd15-441c-a967-ae04ca36a7e6 · outbound

This paper cites Attriguard: Defeating indirect prompt injection in LLM agents via causal attribution of tool invocations.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Attriguard: Defeating indirect prompt injection in LLM agents via causal attribution of tool invocations

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.583531Z

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 e8380c2b-1bff-4d2a-aa42-494dbe876702 · outbound

This paper cites Internal representations as indicators of hallucinations in agent tool selection.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Internal representations as indicators of hallucinations in agent tool selection

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.342044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 857e5167-3ac0-41a7-a5e7-a6535be22dec · outbound

This paper cites Defending Against Indirect Prompt Injection Attacks With Spotlighting.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Defending Against Indirect Prompt Injection Attacks With Spotlighting

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.344108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b1909c9-be26-4bba-9d26-689ae0a87232 · outbound

This paper cites Hsu, and Pin-Yu Chen.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Hsu, and Pin-Yu Chen

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.576749Z

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 753b8d18-880a-439d-9d6c-5efc25da638b · outbound

This paper cites Safepath: Preventing harmful reasoning in chain-of-thought via early alignment.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Safepath: Preventing harmful reasoning in chain-of-thought via early alignment

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.569753Z

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 20cdfd49-d376-474e-a572-fa398ec22ad7 · outbound

This paper cites Llm internal states reveal hallucination risk faced with a query.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Llm internal states reveal hallucination risk faced with a query

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.562396Z

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 6ba31737-4ee0-47ed-b5f0-f41c446651a3 · outbound

This paper cites The task shield: Enforcing task alignment to defend against indirect prompt injection in llm agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure The task shield: Enforcing task alignment to defend against indirect prompt injection in llm agents

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.555511Z

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 09de3710-9633-48d0-9eb8-04646cbf2200 · outbound

This paper cites Promptlocate: Localizing prompt injection attacks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Promptlocate: Localizing prompt injection attacks

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.548137Z

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 2fc9f139-ca42-42a9-9e5a-391b474f2f18 · outbound

This paper cites Activation oracles: Training and evaluating llms as general-purpose activation explainers.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Activation oracles: Training and evaluating llms as general-purpose activation explainers

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.357115Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c531c47a-d446-4eb7-b2a2-98407dffcc16 · outbound

This paper cites Le, and Tomas Pfister.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Le, and Tomas Pfister

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.540719Z

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 132747cf-e63c-4979-82ce-86949a8c70b2 · outbound

This paper cites Hendryx, Summer Yue, and Zifan Wang.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Hendryx, Summer Yue, and Zifan Wang

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.533495Z

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 9c11e71a-89f3-4c64-94a6-067cddbbacce · outbound

This paper cites OS -harm: A benchmark for measuring safety of computer use agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure OS -harm: A benchmark for measuring safety of computer use agents

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.526544Z

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 5aaea5ae-c316-4783-9714-1b9170c73a5a · outbound

This paper cites Measuring AI ability to complete long software tasks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Measuring AI ability to complete long software tasks

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.519743Z

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 bbed69b4-bd07-4688-a003-e43f2c3f5d5d · outbound

This paper cites Efficient memory management for large language model serving with pagedattention.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Efficient memory management for large language model serving with pagedattention

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.512771Z

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 e7c7ba09-adf7-4d49-99a4-c14604383937 · outbound

This paper cites Measuring Faithfulness in Chain-of-Thought Reasoning.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Measuring Faithfulness in Chain-of-Thought Reasoning

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.369612Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39a8ec91-4374-48f9-80f7-73e24dbbf8c2 · outbound

This paper cites Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.371881Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f42caaef-7e1f-422e-8307-c64c9ca00cbe · outbound

This paper cites Inference-time intervention: Eliciting truthful answers from a language model.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Inference-time intervention: Eliciting truthful answers from a language model

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.505775Z

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 6fcfaa42-5014-467e-adb2-77b49e572752 · outbound

This paper cites When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents

Reference 50

Resolution: verified exact; local_arxiv, observed 2026-08-05T00:54:57.900478Z

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 e3e69242-191e-4971-bbae-f7de1d05b9dc · outbound

This paper cites Eia: Environmental injection attack on generalist web agents for privacy leakage.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Eia: Environmental injection attack on generalist web agents for privacy leakage

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.499125Z

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 a21ab69d-9501-4b4d-8754-cdc997fe6bdd · outbound

This paper cites Vigil: Defending llm agents against tool stream injection via verify-before-commit.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Vigil: Defending llm agents against tool stream injection via verify-before-commit

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.380837Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58686ce0-c64a-498a-bb7f-8efaaba781bf · outbound

This paper cites SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.382991Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05a13b9a-444d-4532-bbee-ba60d073641f · outbound

This paper cites Traceaegis: Securing llm-based agents via hierarchical and behavioral anomaly detection.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Traceaegis: Securing llm-based agents via hierarchical and behavioral anomaly detection

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.385335Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0979778b-760e-4726-baed-2337ad6f4e30 · outbound

This paper cites Prompt Injection attack against LLM-integrated Applications.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Prompt Injection attack against LLM-integrated Applications

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.388047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b178dcf-91be-47ec-9a6d-c6d8c194ea2f · outbound

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

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Formalizing and benchmarking prompt injection attacks and defenses

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.492224Z

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 71eee48b-2e5a-47dc-bf87-71ffb9f381d7 · outbound

This paper cites Bogdan, Senthooran Rajamanoharan, and Neel Nanda.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Bogdan, Senthooran Rajamanoharan, and Neel Nanda

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.485385Z

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 c4d8cee7-3389-4157-b64f-ca4ec9ef9547 · outbound

This paper cites s1: Simple test-time scaling.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure s1: Simple test-time scaling

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.478348Z

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 893b2cf9-786d-4bd2-aa9f-68352e604a15 · outbound

This paper cites The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.396764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c718d3d1-ec12-45bf-b72d-60bea70b0055 · outbound

This paper cites GPT-5.6 System Card.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure GPT-5.6 System Card

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.471150Z

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 04037920-c5bb-4d12-9836-f7b1aad4e620 · outbound

This paper cites GPT‑5.5 System Card.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure GPT‑5.5 System Card

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.464321Z

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 e515aa37-cf34-42f9-8805-3b0760dd2846 · outbound

This paper cites LLM s know more than they show: On the intrinsic representation of LLM hallucinations.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure LLM s know more than they show: On the intrinsic representation of LLM hallucinations

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.457553Z

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 9a0ba171-7d4a-472c-9b56-b401a90e7323 · outbound

This paper cites Owasp top 10 for agentic applications for 2026, 2025.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Owasp top 10 for agentic applications for 2026, 2025

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.450194Z

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 d6bff5c4-1b84-4aa0-81c9-a23ebc63fc4a · outbound

This paper cites Latent QA : Teaching LLM s to decode activations into natural language.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Latent QA : Teaching LLM s to decode activations into natural language

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.443308Z

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 78ca08a7-dca1-4c8f-89db-4590e1a59204 · outbound

This paper cites The linear representation hypothesis and the geometry of large language models.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure The linear representation hypothesis and the geometry of large language models

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.409979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d03875ff-f35d-4ec4-98f4-7b63a190030b · outbound

This paper cites Scikit-learn: Machine learning in python.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Scikit-learn: Machine learning in python

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.412072Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd11d624-c7f0-4fc0-97cc-22b34db60500 · outbound

This paper cites Ignore previous prompt: Attack techniques for language models.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Ignore previous prompt: Attack techniques for language models

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.427292Z

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 5d5ee724-8cc8-41e5-ae4f-6fe05c7b9439 · outbound

This paper cites Steering llama 2 via contrastive activation addition.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Steering llama 2 via contrastive activation addition

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.420169Z

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 86e8a1cb-7287-4fce-9297-bc8354c6458c · outbound

This paper cites Great, now write an article about that: The crescendo multi-turn llm jailbreak attack.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Great, now write an article about that: The crescendo multi-turn llm jailbreak attack

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.412600Z

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 863184d2-8aa0-42be-814d-e1743bc4b8c1 · outbound

This paper cites Ignore this title and hackaprompt: Exposing systemic vulnerabilities of llms through a global prompt hacking competition.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Ignore this title and hackaprompt: Exposing systemic vulnerabilities of llms through a global prompt hacking competition

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.404536Z

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 1f4cf12f-5893-4691-8b22-b7ef8e271ac8 · outbound

This paper cites PromptArmor: Simple yet Effective Prompt Injection Defenses.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure PromptArmor: Simple yet Effective Prompt Injection Defenses

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.422665Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6472a7d6-1abf-4bdd-9660-c053fd77c4d3 · outbound

This paper cites Daniel Freeman, Theodore R.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Daniel Freeman, Theodore R

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.424978Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 732b4a58-e52c-4901-aefa-6e93afcece8d · outbound

This paper cites Tensor trust: Interpretable prompt injection attacks from an online game.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Tensor trust: Interpretable prompt injection attacks from an online game

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.393335Z

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 304d6cf6-3ab3-4819-bb76-258b988ee964 · outbound

This paper cites The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.429487Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 79e606ba-5f75-47a2-b994-d20998f4f783 · outbound

This paper cites Raccoon: Prompt extraction benchmark of llm-integrated applications.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Raccoon: Prompt extraction benchmark of llm-integrated applications

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.386468Z

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 8f6737ba-73f0-4be0-a105-0ee6ecf82c30 · outbound

This paper cites Agentarmor: Enforcing program analysis on agent runtime trace to defend against prompt injection.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Agentarmor: Enforcing program analysis on agent runtime trace to defend against prompt injection

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.434034Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86c19c39-f6c1-4edf-83cc-2e23a301e48b · outbound

This paper cites Automatic layer selection for hallucination detection.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Automatic layer selection for hallucination detection

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.379900Z

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 15774ad1-a8ea-4106-b132-d3bab8aa8cfd · outbound

This paper cites Defending against indirect prompt injection by instruction detection.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Defending against indirect prompt injection by instruction detection

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.373573Z

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 262a2156-1068-4ed9-8947-a7e8191de3eb · outbound

This paper cites an unresolved cited work.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Unresolved cited work

Reference 79

Resolution: unresolved; raw_fallback, observed 2026-08-05T00:54:58.367004Z

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 6ef89cfa-42d3-4f04-9db6-015860dea1f5 · outbound

This paper cites Guardagent: Safeguard LLM agents via knowledge-enabled reasoning.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Guardagent: Safeguard LLM agents via knowledge-enabled reasoning

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.360303Z

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 1d493fda-8275-47c3-aafe-268b4c406d5e · outbound

This paper cites Qwen3 Technical Report.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Qwen3 Technical Report

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.447210Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6fe9ac39-009d-4c96-9856-8ad2cf964fa3 · outbound

This paper cites RAP-ID : Mechanistic prompt injection detection via impostor behavior analysis.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure RAP-ID : Mechanistic prompt injection detection via impostor behavior analysis

Reference 82

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.353289Z

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 a53c2fca-6392-45a8-b876-3621dc0f40b5 · outbound

This paper cites The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.452102Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fba3050a-4673-47d2-9efc-7f027e406c2d · outbound

This paper cites Injecagent: Benchmarking indirect prompt injections in tool-integrated large language model agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Injecagent: Benchmarking indirect prompt injections in tool-integrated large language model agents

Reference 84

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.346219Z

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 345000b5-0375-4133-b06d-bbf6b52daf2a · outbound

This paper cites Adaptive attacks break defenses against indirect prompt injection attacks on llm agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Adaptive attacks break defenses against indirect prompt injection attacks on llm agents

Reference 85

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.339017Z

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 1b723bfd-acaa-4072-bbc1-098d825e287e · outbound

This paper cites Agent security bench ( ASB ): Formalizing and benchmarking attacks and defenses in LLM -based agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Agent security bench ( ASB ): Formalizing and benchmarking attacks and defenses in LLM -based agents

Reference 86

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.331177Z

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 6f700c98-420c-45a9-ab76-da7a81683c1a · outbound

This paper cites Defense against prompt injection attacks via mixture of encodings.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Defense against prompt injection attacks via mixture of encodings

Reference 87

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.323879Z

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 5244c1af-d5fe-4fe5-93b3-f3e7a6e68f39 · outbound

This paper cites Toolbehonest: A multi-level hallucination diagnostic benchmark for tool-augmented large language models.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Toolbehonest: A multi-level hallucination diagnostic benchmark for tool-augmented large language models

Reference 88

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.316206Z

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 ce8c1d43-1646-4b00-85f2-cf3c0e528993 · outbound

This paper cites Iheval: Evaluating language models on following the instruction hierarchy.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Iheval: Evaluating language models on following the instruction hierarchy

Reference 89

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.308618Z

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 21c06b96-d2df-4fde-88d3-b7fe379aefa8 · outbound

This paper cites Attention is all you need to defend against indirect prompt injection attacks in llms.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Attention is all you need to defend against indirect prompt injection attacks in llms

Reference 90

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.301434Z

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 8223654b-2157-406e-b45c-b8ecaefc50dd · outbound

This paper cites MELON : Provable defense against indirect prompt injection attacks in AI agents.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure MELON : Provable defense against indirect prompt injection attacks in AI agents

Reference 91

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T00:54:58.292845Z

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 1ff8edec-df0e-4de2-a9c9-0cec8a9caa21 · outbound

This paper cites Your Agent is More Brittle Than You Think: Uncovering Indirect Injection Vulnerabilities in Agentic LLMs.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Your Agent is More Brittle Than You Think: Uncovering Indirect Injection Vulnerabilities in Agentic LLMs

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.474502Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb73bb13-2cdf-4042-afa8-7ed1493e80ed · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Representation Engineering: A Top-Down Approach to AI Transparency

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.476993Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2b6154e-2b53-4867-94d6-4e6d841655d0 · outbound

This paper cites Pishield: Detecting prompt injection attacks via intrinsic llm features.

Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure Pishield: Detecting prompt injection attacks via intrinsic llm features

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-05T00:54:57.479602Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers