Paper Citation Record · LEDGER

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles

As of 13 August 2026, this Paper Citation Record lists 16 of 16 outbound references and 0 inbound Pith citation observations for arXiv:2607.23379.

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

pith.paper-citation-record.v1
2607.23379 v1

Coverage vector

measured 16 of 16 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-30T23:56:50.982582Z

measured 16 of 16 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-13T06:32:02.005865+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

16 of 16 outbound references displayed

verified exact0
verified fuzzy0
unresolved15
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 7f8b8bc4-290d-4f38-903d-19d6529f4a1c · outbound

This paper cites Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers , journal =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers , journal =

Reference 1

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.926385Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c66e534-759f-41f8-a162-f1e3bc12537b · outbound

This paper cites Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences , journal =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences , journal =

Reference 2

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.930261Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df4e3907-474c-4909-9ba9-029b28afc797 · outbound

This paper cites Towards eliciting latent knowledge from LLMs with mechanistic interpretability.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Towards eliciting latent knowledge from LLMs with mechanistic interpretability

Reference 3

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.935284Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca885bb5-326f-482c-b284-ff597d907a01 · outbound

This paper cites 2020 , month = aug, howpublished =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles 2020 , month = aug, howpublished =

Reference 4

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.939643Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 525845d8-95e9-41c1-9dcb-ed7680a2d315 · outbound

This paper cites Manning , editor =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Manning , editor =

Reference 5

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.943120Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18108224-c808-45a4-bd40-138a25abcc55 · outbound

This paper cites Scaling and evaluating sparse autoencoders , booktitle =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Scaling and evaluating sparse autoencoders , booktitle =

Reference 6

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.947020Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3a7be2c-0fd3-4dd6-9f3c-01b7e6380be6 · outbound

This paper cites The Fourteenth International Conference on Learning Representations , year=.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles The Fourteenth International Conference on Learning Representations , year=

Reference 7

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.950667Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fa56c4d8-cdb3-42b9-99d8-29df3d585f40 · outbound

This paper cites Advances in Neural Information Processing Systems , year =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Advances in Neural Information Processing Systems , year =

Reference 8

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.953964Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49d56262-fd14-439d-900e-3342b942b3be · outbound

This paper cites 2026 , howpublished =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles 2026 , howpublished =

Reference 9

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.957240Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f087f9c3-1159-4b8d-af6e-18f874b2b582 · outbound

This paper cites 2021 , howpublished =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles 2021 , howpublished =

Reference 10

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.960558Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 713d0b27-5ac5-4134-9904-c464f52fb379 · outbound

This paper cites NeurIPS Workshop on Machine Learning and Computer Security , year =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles NeurIPS Workshop on Machine Learning and Computer Security , year =

Reference 11

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.964942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 546c1967-a201-4e8b-bf37-840eb7df9bc2 · outbound

This paper cites IEEE Symposium on Security and Privacy , year =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles IEEE Symposium on Security and Privacy , year =

Reference 12

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.968350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01fd38ed-a2d9-481b-a039-93c97a02448a · outbound

This paper cites BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.971987Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7e174644-e2a3-4a37-84ab-e20e6d0724e0 · outbound

This paper cites an unresolved cited work.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Unresolved cited work

Reference 14

Resolution: malformed identifier; no resolver link, observed 2026-07-30T23:56:50.975706Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f97e2546-f5b8-4fc8-a471-6b46d438a6f2 · outbound

This paper cites The Twelfth International Conference on Learning Representations,.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles The Twelfth International Conference on Learning Representations,

Reference 15

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.979240Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 165aab8f-9e86-4ae9-a99b-e29ca7c2bcca · outbound

This paper cites Enhancing Chat Language Models by Scaling High-quality Instructional Conversations , booktitle =.

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Enhancing Chat Language Models by Scaling High-quality Instructional Conversations , booktitle =

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-30T23:56:50.982582Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.