Paper Citation Record · LEDGER

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations

As of 7 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 2 inbound Pith citation observations for arXiv:2507.23221.

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

pith.paper-citation-record.v1
2507.23221 v1

Coverage vector

measured 37 of 37 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T11:01:15.053772Z

measured 39 of 39 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-21T21:44:36.351517Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-21T21:45:40.587851Z

Reference resolution

37 of 37 outbound references displayed

verified exact1
verified fuzzy5
unresolved31
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 25dac620-b13b-4dcb-80fe-451c71a158d6 · outbound

This paper cites Understanding intermediate layers using linear classifier probes.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Understanding intermediate layers using linear classifier probes

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.932935Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3d95ff1-40eb-4a7a-905e-475f22780c93 · outbound

This paper cites and Mitchell, T.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations and Mitchell, T

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.936933Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f213f5fe-6f27-4bc3-beb0-5c38c1498002 · outbound

This paper cites Discovering Latent Knowledge in Language Models Without Supervision.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Discovering Latent Knowledge in Language Models Without Supervision

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T11:01:15.584421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 7c804f6f-7361-4ebe-bee7-6c1e988777e3 · outbound

This paper cites an unresolved cited work.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Unresolved cited work

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.943561Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aca8bfaa-8d87-483a-b4c1-a03cdfa3dcf1 · outbound

This paper cites Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.946851Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d8f589a-a897-48c2-8201-98f30091193a · outbound

This paper cites Sparse Autoencoders Find Highly Interpretable Features in Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.950678Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bae3f20a-b573-41b3-837c-3d90c4fbcc55 · outbound

This paper cites Toy Models of Superposition.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Toy Models of Superposition

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.954567Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ee00304-7644-481d-92c9-60ec99b60ba5 · outbound

This paper cites J., Gurnee, W., and Tegmark, M.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations J., Gurnee, W., and Tegmark, M

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T11:01:15.573589Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation cf51700b-ea7a-45f0-8302-84d492486e92 · outbound

This paper cites Detecting hallucinations in large language models using semantic entropy.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Detecting hallucinations in large language models using semantic entropy

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.961171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99c7fa8d-dbdd-45da-8611-e9028fd05045 · outbound

This paper cites Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.964076Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8775890-8743-4f70-bf32-4465dae99fd6 · outbound

This paper cites The Pile: An 800GB Dataset of Diverse Text for Language Modeling.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations The Pile: An 800GB Dataset of Diverse Text for Language Modeling

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.967476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 010c7cde-f373-46aa-947e-2f89ad41d465 · outbound

This paper cites spacy: Industrial-strength natural language processing in python.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations spacy: Industrial-strength natural language processing in python

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T11:01:15.563539Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation e448f1d3-c66a-4869-91b7-84c252dbd04e · outbound

This paper cites A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.973911Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f853e8fc-909e-401a-818b-d4bb84f6455e · outbound

This paper cites Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.977196Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 316b27aa-2979-4ab1-bcc6-425a3813d766 · outbound

This paper cites Survey of Hallucination in Natural Language Generation.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Survey of Hallucination in Natural Language Generation

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.980382Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8aedb19d-a45e-4668-b3dc-72ada4946a1f · outbound

This paper cites Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.983343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2ffc54d-ff52-4ec8-9a4c-1eca17ef4fb1 · outbound

This paper cites k-Sparse Autoencoders.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations k-Sparse Autoencoders

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.986525Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9061660b-fc77-436e-94c7-068de140ee8a · outbound

This paper cites an unresolved cited work.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Unresolved cited work

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.989995Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ea7fb9a-8e14-4c15-8014-6f4e2fe1fda6 · outbound

This paper cites Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.993002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e78388c6-ec3c-42d3-8241-5bd3ed2b75d5 · outbound

This paper cites On Faithfulness and Factuality in Abstractive Summarization.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations On Faithfulness and Factuality in Abstractive Summarization

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.996355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f3a57042-15f0-415e-ae1b-0d9ce27ab1a2 · outbound

This paper cites Controllable Context Sensitivity and the Knob Behind It.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Controllable Context Sensitivity and the Knob Behind It

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:14.999648Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99f8c185-00f4-4cde-8bf2-6a90e1ca62a0 · outbound

This paper cites A., and Kriegeskorte, N.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations A., and Kriegeskorte, N

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T11:01:15.552912Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 010d2a4c-07c6-495d-8649-4533985454be · outbound

This paper cites Entity-level Factual Consistency of Abstractive Text Summarization.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Entity-level Factual Consistency of Abstractive Text Summarization

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.005942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 575fcf43-1924-448c-a64e-02a093430aec · outbound

This paper cites Emergent Linear Representations in World Models of Self-Supervised Sequence Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Emergent Linear Representations in World Models of Self-Supervised Sequence Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.009218Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18e6b3a5-e525-4361-afe7-0f8cc9432132 · outbound

This paper cites Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.012396Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd9dcc5f-e8d4-4cfd-a90b-f22ad9631edf · outbound

This paper cites The Linear Representation Hypothesis and the Geometry of Large Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations The Linear Representation Hypothesis and the Geometry of Large Language Models

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.015631Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2bc2c98d-2c5c-4c7f-b88b-53efa68d1e78 · outbound

This paper cites A practical review of mechanistic interpretability for transformer-based language models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations A practical review of mechanistic interpretability for transformer-based language models

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.018933Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b84aaaa-522b-4386-bbfc-6f33d8465876 · outbound

This paper cites Hallushield: A mechanistic approach to hallucination resistant models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Hallushield: A mechanistic approach to hallucination resistant models

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T11:01:15.542375Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 44e1a28c-b63a-40a4-a831-d01c2428d807 · outbound

This paper cites Get To The Point: Summarization with Pointer-Generator Networks.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Get To The Point: Summarization with Pointer-Generator Networks

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.025985Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e60da72d-a791-43c5-92f8-e6bdf59ff759 · outbound

This paper cites Constructing Benchmarks and Interventions for Combating Hallucinations in LLMs.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Constructing Benchmarks and Interventions for Combating Hallucinations in LLMs

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.029328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29f924ff-d768-4776-ab16-9827381fb2ac · outbound

This paper cites Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.033088Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 447bea56-d038-44d7-9624-3c81ac77f357 · outbound

This paper cites The Curious Case of Hallucinatory (Un)answerability: Finding Truths in the Hidden States of Over-Confident Large Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations The Curious Case of Hallucinatory (Un)answerability: Finding Truths in the Hidden States of Over-Confident Large Language Models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.036213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0ba1cff9-5cf9-4d5a-9248-59565109c9fc · outbound

This paper cites Redeep: Detecting hallucination in retrieval augmented generation via mechanistic interpretability.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Redeep: Detecting hallucination in retrieval augmented generation via mechanistic interpretability

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.039601Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2d009fdf-20bb-4135-8758-51881219073e · outbound

This paper cites Cost-Effective Hallucination Detection for LLMs.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Cost-Effective Hallucination Detection for LLMs

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-08-06T11:01:15.147227Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation c43006d9-345a-43c0-b362-fbd2b37b9da3 · outbound

This paper cites an unresolved cited work.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Unresolved cited work

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.045482Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 59842f00-0487-43b2-9205-a8bc4677d5e6 · outbound

This paper cites Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.050200Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbcffd29-bb0b-4789-b68e-c050bb0b1f89 · outbound

This paper cites write newline.

A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations write newline

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-06T11:01:15.053772Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 8d0a9c56-e1aa-4ea9-9116-4297c04a8041 · inbound

Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models cites this paper.

Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-21T21:45:40.590400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 4097f8f0-7c55-4ee8-97d6-95ccbfba6437 · inbound

PARALLAX: Separating Genuine Hallucination Detection from Benchmark Construction Artifacts cites this paper.

PARALLAX: Separating Genuine Hallucination Detection from Benchmark Construction Artifacts A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations

Reference 51

Resolution: metadata mismatch; arxiv_id, observed 2026-05-19T20:12:44.923519Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.