Paper Citation Record · LEDGER

Because we have LLMs, we Can and Should Pursue Agentic Interpretability

As of 15 August 2026, this Paper Citation Record lists 28 of 28 outbound references and 6 inbound Pith citation observations for arXiv:2506.12152.

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

pith.paper-citation-record.v1
2506.12152 v1

Coverage vector

measured 28 of 28 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T01:03:22.328312Z

measured 34 of 34 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-15T06:32:42.880941+00:00

measured 6 of 6 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-03T03:50:15.977348Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-04T00:59:21.346717Z

Reference resolution

28 of 28 outbound references displayed

verified exact1
verified fuzzy2
unresolved24
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 91d74f30-bc4c-459d-9d54-e29fb3303496 · outbound

This paper cites Abdul, J.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Abdul, J

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T01:03:23.670680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 7309861d-5495-4868-847b-d9a5f7b77729 · outbound

This paper cites Studying Large Language Model Generalization with Influence Functions.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Studying Large Language Model Generalization with Influence Functions

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.807887Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5660c1af-c3fc-4483-a960-a1a2a1234d4b · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.917665Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 065723a3-5382-48fc-8277-e83830fc674f · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.093735Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99aaa16c-8ae6-4aa2-b35e-a55530ecda6a · outbound

This paper cites doi: 10.1145/3564240.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability doi: 10.1145/3564240

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.338690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b20b3980-7f26-480f-9757-23a47c91446e · outbound

This paper cites An Approach to Technical AGI Safety and Security.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability An Approach to Technical AGI Safety and Security

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.407547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5584384c-eafb-41cf-9f8c-642c945093dd · outbound

This paper cites Open Problems in Mechanistic Interpretability.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Open Problems in Mechanistic Interpretability

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.495743Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2e9c52c-864f-43da-97e9-fc5dcca42dd8 · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.637308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7bdb23f-159c-4c85-a2eb-c10b77497794 · outbound

This paper cites doi: 10.18653/v1/D16-1159.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability doi: 10.18653/v1/D16-1159

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.710057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb7f58e0-4da1-417f-8b70-2b500455401a · outbound

This paper cites Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.917517Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 41a21ca5-2781-41f9-a691-92e990be6664 · outbound

This paper cites A Roadmap to Pluralistic Alignment.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability A Roadmap to Pluralistic Alignment

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.990940Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9a5a264-c583-455f-8512-4cb5075a344e · outbound

This paper cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:22.291884Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9a5d7c8-0322-4597-9f3f-d296e7bb0d50 · outbound

This paper cites Large Language Models Are Human-Level Prompt Engineers.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Large Language Models Are Human-Level Prompt Engineers

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:22.328312Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9c4d9f5-7700-4a11-892a-e4fe517619da · outbound

This paper cites URLhttps://www.ncbi.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability URLhttps://www.ncbi

Reference 1974

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:22.067758Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28924536-ca9b-4f84-a1e7-c13635f86df4 · outbound

This paper cites Merriam-webster, Accessed on 2025-05-22.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Merriam-webster, Accessed on 2025-05-22

Reference 1982

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T01:03:23.233742Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation a2856cf9-3960-4f23-8784-086a987b8959 · outbound

This paper cites Designing a Dashboard for Transparency and Control of Conversational AI.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Designing a Dashboard for Transparency and Control of Conversational AI

Reference 1993

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.611033Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ecd4a1a0-0a54-48aa-9cdd-44aae1d59611 · outbound

This paper cites Alignment faking in large language models.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Alignment faking in large language models

Reference 2012

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.741567Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bdaa9f40-15d2-4727-9204-0c22b5514aad · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Constitutional AI: Harmlessness from AI Feedback

Reference 2014

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.338834Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 402b051c-2f8b-494d-9fe3-1b581854a724 · outbound

This paper cites doi: 10.18653/v1/W16-2524.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability doi: 10.18653/v1/W16-2524

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.677305Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f079495-60a1-40d6-a715-75f588fa91e0 · outbound

This paper cites Auditing language models for hidden objectives.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Auditing language models for hidden objectives

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.160673Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c08c3d79-8d55-437c-8cf4-6c9365c936ab · outbound

This paper cites Sampling Method for Fast Training of Support Vector Data Description.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Sampling Method for Fast Training of Support Vector Data Description

Reference 2018

Resolution: metadata mismatch; local_arxiv, observed 2026-08-07T01:03:23.104484Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 56ab1b47-fcb6-4cfd-99c8-6d4a8185974d · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 2019

Resolution: unresolved; raw_fallback, observed 2026-08-07T01:03:23.350522Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 8f150441-741f-4da9-b5ca-13a1248d9ad8 · outbound

This paper cites AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.800720Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6adca01-aeda-4876-9a8b-2b6856c8ceec · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 2021

Resolution: verified exact; doi, observed 2026-08-07T01:03:22.527682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation c147767b-17da-4232-8d06-db94f7cb54d5 · outbound

This paper cites an unresolved cited work.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Unresolved cited work

Reference 2022

Resolution: unresolved; raw_fallback, observed 2026-08-07T01:03:23.491721Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 009c88f1-a1f1-46ef-8ef4-504a046f3b1c · outbound

This paper cites Progress measures for grokking via mechanistic interpretability.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Progress measures for grokking via mechanistic interpretability

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.272090Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a287e320-3fd1-421a-b9ef-8b7bf845bac5 · outbound

This paper cites Challenges in Human-Agent Communication.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability Challenges in Human-Agent Communication

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:20.470352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2bd2d3f5-d96d-47ae-85b0-ee97be0b6d7f · outbound

This paper cites We Can't Understand AI Using our Existing Vocabulary.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability We Can't Understand AI Using our Existing Vocabulary

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-07T01:03:21.013201Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 8949abfc-dd1c-4b41-85b6-5aed53fdaacc · inbound

Adaptive Chain-of-Focus Reasoning via Dynamic Visual Search and Zooming for Efficient VLMs cites this paper.

Adaptive Chain-of-Focus Reasoning via Dynamic Visual Search and Zooming for Efficient VLMs Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-17T05:35:13.274940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 715bf029-5944-4d4b-a090-5a2e8a47486a · inbound

From Features to Actions: Explainability in Traditional and Agentic AI Systems cites this paper.

From Features to Actions: Explainability in Traditional and Agentic AI Systems Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T03:50:15.977348Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9b42aa7-1170-456f-92e8-9d6809daa34b · inbound

A Geometric View for Understanding Concept Learning and Neuron Interpretation in Sparse Autoencoders cites this paper.

A Geometric View for Understanding Concept Learning and Neuron Interpretation in Sparse Autoencoders Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 55

Resolution: metadata mismatch; arxiv_id, observed 2026-07-02T16:47:09.928488Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation d0440bd1-9207-4522-b016-ff15cda7a584 · inbound

Uncertainty Decomposition for Clarification Seeking in LLM Agents cites this paper.

Uncertainty Decomposition for Clarification Seeking in LLM Agents Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 14

Resolution: metadata mismatch; arxiv_id, observed 2026-07-04T00:59:21.349255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 6e828df5-a319-45ff-a816-957d22a9a462 · inbound

The Curse of Multiple Mediators: Hidden Interaction Effects in Activation Patching cites this paper.

The Curse of Multiple Mediators: Hidden Interaction Effects in Activation Patching Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-07-01T18:45:58.488988Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 78706c2d-689e-48bf-a2ca-7c96dc239c67 · inbound

Reality Monitoring in Large Language Models: Self-Knowledge That Transforms with Conversation Memory cites this paper.

Reality Monitoring in Large Language Models: Self-Knowledge That Transforms with Conversation Memory Because we have LLMs, we Can and Should Pursue Agentic Interpretability

Reference 72

Resolution: unresolved; no resolver link, observed 2026-07-31T23:35:49.490635Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.