Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-05-17T20:26:37.914522Z
Paper Citation Record · LEDGER
The Impact of Off-Policy Training Data on Probe Generalisation
As of 4 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 1 inbound Pith citation observation for arXiv:2511.17408.
A citation records a reference. It does not transfer a finding from one paper to another.
Coverage vector
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links, observed 2026-07-30T21:44:25.201514Z
A source-named dated measurement, never combined with another source.
Source: cited_works
Reference resolution
45 of 45 outbound references displayed
- verified exact25
- verified fuzzy7
- unresolved0
- parse uncertain0
- malformed identifier0
- metadata mismatch13
External citation measurements
No source-named external measurement is stored.
Outbound references
Observation 4a8a3a90-05a4-40f9-b188-114f64780575 · outbound
This paper cites Linear Control of Test Awareness Reveals Differential Compliance in Reasoning Models , May 2025.
The Impact of Off-Policy Training Data on Probe Generalisation Linear Control of Test Awareness Reveals Differential Compliance in Reasoning Models , May 2025
Reference 1
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.631056Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c4ee3199-3bb7-4bf7-a877-572424067f4d · outbound
This paper cites Invariant Risk Minimization.
The Impact of Off-Policy Training Data on Probe Generalisation Invariant Risk Minimization
Reference 2
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.624218Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a0976b27-f64c-4805-9076-84da2d039a73 · outbound
This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.
The Impact of Off-Policy Training Data on Probe Generalisation Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
Reference 3
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.683687Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a909a741-651d-4fb4-8881-87be77fe645a · outbound
This paper cites Deception in LLMs: Self-Preservation and Autonomous Goals in Large Language Models.
The Impact of Off-Policy Training Data on Probe Generalisation Deception in LLMs: Self-Preservation and Autonomous Goals in Large Language Models
Reference 4
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.686788Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 6c0a7faa-e46b-40ab-bb19-68de55c9c107 · outbound
This paper cites Sabotage Evaluations for Frontier Models.
The Impact of Off-Policy Training Data on Probe Generalisation Sabotage Evaluations for Frontier Models
Reference 5
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.717804Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 82feaba0-7f2d-4d2e-b478-82de071c8d7a · outbound
This paper cites Propositional Interpretability in Artificial Intelligence.
The Impact of Off-Policy Training Data on Probe Generalisation Propositional Interpretability in Artificial Intelligence
Reference 6
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.678096Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 0909cc93-57ac-4aaf-a3a8-5145ba3e81bd · outbound
This paper cites month = jul, year =.
The Impact of Off-Policy Training Data on Probe Generalisation month = jul, year =
Reference 7
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.720919Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 0a07bec9-ee27-4f4c-89cc-d95e5b19e0d5 · outbound
This paper cites Cost-effective constitutional classifiers via representation re-use.
The Impact of Off-Policy Training Data on Probe Generalisation Cost-effective constitutional classifiers via representation re-use
Reference 8
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.634244Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d7b924d5-dd2e-415e-83b2-c5c32941f2bf · outbound
This paper cites DeepSeek-V3 Technical Report.
The Impact of Off-Policy Training Data on Probe Generalisation DeepSeek-V3 Technical Report
Reference 9
- Resolution
- metadata mismatch
- local_arxiv, observed 2026-05-17T20:30:11.723980Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a002ea0c-53d2-4133-9796-41a052437ee4 · outbound
This paper cites Enhancing Chat Language Models by Scaling High-quality Instructional Conversations.
The Impact of Off-Policy Training Data on Probe Generalisation Enhancing Chat Language Models by Scaling High-quality Instructional Conversations
Reference 10
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.731506Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7aa3136f-6cee-4b02-b2ee-c536505dc6db · outbound
This paper cites Hierarchical Neural Story Generation.
The Impact of Off-Policy Training Data on Probe Generalisation Hierarchical Neural Story Generation
Reference 11
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.680944Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 959b5e21-c011-42e8-b79d-07922309fb87 · outbound
This paper cites Monitoring Latent World States in Language Models with Propositional Probes.
The Impact of Off-Policy Training Data on Probe Generalisation Monitoring Latent World States in Language Models with Propositional Probes
Reference 12
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.741041Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a94f196f-6b16-4e25-af5c-d6b9431a1a50 · outbound
This paper cites Detecting Strategic Deception Using Linear Probes.
The Impact of Off-Policy Training Data on Probe Generalisation Detecting Strategic Deception Using Linear Probes
Reference 13
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.634956Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 02cf7c9a-d832-48d7-b219-06c3e5ffc297 · outbound
This paper cites Probing the Robustness of Large Language Models Safety to Latent Perturbations.
The Impact of Off-Policy Training Data on Probe Generalisation Probing the Robustness of Large Language Models Safety to Latent Perturbations
Reference 14
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.710099Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d3bfb3de-ec41-4833-afea-76c02aa7920b · outbound
This paper cites What makes a convincing argument? Empirical analysis and detecting attributes of convincingness in Web argumentation.
The Impact of Off-Policy Training Data on Probe Generalisation What makes a convincing argument? Empirical analysis and detecting attributes of convincingness in Web argumentation
Reference 15
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.631651Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 79ac2412-2d6f-42c6-b198-7b381d400692 · outbound
This paper cites Measuring Massive Multitask Language Understanding.
The Impact of Off-Policy Training Data on Probe Generalisation Measuring Massive Multitask Language Understanding
Reference 16
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.751114Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation bd0d3745-0d29-4e4f-b4ab-99c548b7221d · outbound
This paper cites Ministral-8b-instruct-2410 model card.
The Impact of Off-Policy Training Data on Probe Generalisation Ministral-8b-instruct-2410 model card
Reference 17
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.636187Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 1eb8a53c-45f6-4076-b224-cb1d80a8e7b5 · outbound
This paper cites Mistral 7B.
The Impact of Off-Policy Training Data on Probe Generalisation Mistral 7B
Reference 18
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.734593Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 25a73da2-530c-4daa-ab37-c870c76670f1 · outbound
This paper cites Mixtral of Experts.
The Impact of Off-Policy Training Data on Probe Generalisation Mixtral of Experts
Reference 19
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.696061Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 1b81dfbf-c725-44ca-b6bf-bb259fa14fd8 · outbound
This paper cites WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models.
The Impact of Off-Policy Training Data on Probe Generalisation WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models
Reference 20
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.690062Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ca1cfab4-bcff-43a4-87ba-f32f77a1ec62 · outbound
This paper cites HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden States.
The Impact of Off-Policy Training Data on Probe Generalisation HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden States
Reference 21
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.638926Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 70b657d8-d71f-47a7-bc9c-175570f96ead · outbound
This paper cites What Features in Prompts Jailbreak LLMs ? Investigating the Mechanisms Behind Attacks , May 2025.
The Impact of Off-Policy Training Data on Probe Generalisation What Features in Prompts Jailbreak LLMs ? Investigating the Mechanisms Behind Attacks , May 2025
Reference 22
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.693163Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 92aa5abd-c7f2-492e-b7c8-0e560de0b304 · outbound
This paper cites The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning.
The Impact of Off-Policy Training Data on Probe Generalisation The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
Reference 23
- Resolution
- metadata mismatch
- local_arxiv, observed 2026-05-17T20:30:11.702463Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 3a8746ac-936c-4f88-acec-c7b24dccc6ad · outbound
This paper cites Simple probes can catch sleeper agents, April 2024.
The Impact of Off-Policy Training Data on Probe Generalisation Simple probes can catch sleeper agents, April 2024
Reference 24
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.628778Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e24a5014-e9cf-435f-93f7-3f24e5e0918a · outbound
This paper cites HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.
The Impact of Off-Policy Training Data on Probe Generalisation HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
Reference 25
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.668116Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ca924539-86f7-4c76-a6dd-40580c9414ac · outbound
This paper cites Detecting.
The Impact of Off-Policy Training Data on Probe Generalisation Detecting
Reference 26
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.713833Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation fefb9d7f-eb7c-40cd-a25d-02dd83fb947e · outbound
This paper cites Llama 3.2: Revolutionizing edge ai and vision (connect 2024).
The Impact of Off-Policy Training Data on Probe Generalisation Llama 3.2: Revolutionizing edge ai and vision (connect 2024)
Reference 27
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.638155Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d7c3d801-dbdc-4dd5-9ac4-4c72dc6b0793 · outbound
This paper cites Probing and Steering Evaluation Awareness of Language Models.
The Impact of Off-Policy Training Data on Probe Generalisation Probing and Steering Evaluation Awareness of Language Models
Reference 28
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.674796Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 01f125d5-1082-46f1-b19c-2b2f1616d802 · outbound
This paper cites Gpt-5 system card.
The Impact of Off-Policy Training Data on Probe Generalisation Gpt-5 system card
Reference 29
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.626578Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation af0a14e1-a378-45be-91e4-1e257955858b · outbound
This paper cites Benchmarking.
The Impact of Off-Policy Training Data on Probe Generalisation Benchmarking
Reference 30
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.748270Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ef8021db-546e-45e2-a846-846b8b0e1d7e · outbound
This paper cites Qwen2.5 Technical Report.
The Impact of Off-Policy Training Data on Probe Generalisation Qwen2.5 Technical Report
Reference 31
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.671475Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 3079e435-eeb2-4085-b96d-83efcdc32f64 · outbound
This paper cites Coup probes: Catching catastrophes with probes trained off-policy, November 2023.
The Impact of Off-Policy Training Data on Probe Generalisation Coup probes: Catching catastrophes with probes trained off-policy, November 2023
Reference 32
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-17T20:32:05.624238Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7ae73266-38ec-4e72-9fd0-ace740a1501b · outbound
This paper cites Large Language Models can Strategically Deceive their Users when Put Under Pressure.
The Impact of Off-Policy Training Data on Probe Generalisation Large Language Models can Strategically Deceive their Users when Put Under Pressure
Reference 33
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.706210Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 9d2b4a8a-7ebf-4385-9d97-733ebfd00c4b · outbound
This paper cites Stress Testing Deliberative Alignment for Anti-Scheming Training , url =.
The Impact of Off-Policy Training Data on Probe Generalisation Stress Testing Deliberative Alignment for Anti-Scheming Training , url =
Reference 34
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.661669Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c1cba9f9-b963-402d-8ddd-a3f1ccecf800 · outbound
This paper cites Towards Understanding Sycophancy in Language Models.
The Impact of Off-Policy Training Data on Probe Generalisation Towards Understanding Sycophancy in Language Models
Reference 35
- Resolution
- metadata mismatch
- local_arxiv, observed 2026-05-17T20:30:11.664946Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7f9011f4-d26a-4d53-ae5d-8758c55f0b39 · outbound
This paper cites Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.
The Impact of Off-Policy Training Data on Probe Generalisation Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming
Reference 36
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T22:20:43.320085Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ea023788-c5a2-4826-a689-1a9d9fcefc7f · outbound
This paper cites Gemma 3 Technical Report.
The Impact of Off-Policy Training Data on Probe Generalisation Gemma 3 Technical Report
Reference 37
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.627465Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 13e64531-f004-44ad-ad14-43ff8a01c0cc · outbound
This paper cites Investigating task-specific prompts and sparse autoencoders for activation monitoring.
The Impact of Off-Policy Training Data on Probe Generalisation Investigating task-specific prompts and sparse autoencoders for activation monitoring
Reference 38
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.699412Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation cc3be326-0d17-47e5-a29b-fbdb71077bad · outbound
This paper cites Diffusion Earth Mover's Distance and Distribution Embeddings.
The Impact of Off-Policy Training Data on Probe Generalisation Diffusion Earth Mover's Distance and Distribution Embeddings
Reference 39
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.655190Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation fe206e23-4a62-4278-9380-06c4394ddcae · outbound
This paper cites Jailbroken: How Does LLM Safety Training Fail?.
The Impact of Off-Policy Training Data on Probe Generalisation Jailbroken: How Does LLM Safety Training Fail?
Reference 40
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.651388Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation f1888f08-cb37-4685-97a3-ca519fe7e19f · outbound
This paper cites AI Sandbagging: Language Models can Strategically Underperform on Evaluations.
The Impact of Off-Policy Training Data on Probe Generalisation AI Sandbagging: Language Models can Strategically Underperform on Evaluations
Reference 41
- Resolution
- verified exact
- arxiv_id, observed 2026-05-17T20:30:11.658351Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 9503af87-6367-4937-aa1b-b0679ca36d34 · outbound
This paper cites Language Models Learn to Mislead Humans via RLHF.
The Impact of Off-Policy Training Data on Probe Generalisation Language Models Learn to Mislead Humans via RLHF
Reference 42
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.646597Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 42fce047-3444-4496-ace5-7163134d68ad · outbound
This paper cites Qwen3 Technical Report.
The Impact of Off-Policy Training Data on Probe Generalisation Qwen3 Technical Report
Reference 43
- Resolution
- verified exact
- local_arxiv, observed 2026-05-17T20:30:11.642685Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 242bea42-9b0c-4cb5-a049-7ed2e177a98a · outbound
This paper cites Uncovering Latent Chain of Thought Vectors in Language Models.
The Impact of Off-Policy Training Data on Probe Generalisation Uncovering Latent Chain of Thought Vectors in Language Models
Reference 44
- Resolution
- metadata mismatch
- arxiv_id, observed 2026-05-17T20:30:11.727486Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d506ccca-0558-4d99-a37f-5f0ee8994d3b · outbound
This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.
The Impact of Off-Policy Training Data on Probe Generalisation Representation Engineering: A Top-Down Approach to AI Transparency
Reference 45
- Resolution
- metadata mismatch
- local_arxiv, observed 2026-05-17T20:30:11.744338Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Pith citing papers
Observation f6d7ce6b-f995-4274-8788-22a11b04f80c · inbound
Two Regimes of Chain-of-Thought Unfaithfulness: Behavioral Detection Fails Where Models Are Wrong cites this paper.
Two Regimes of Chain-of-Thought Unfaithfulness: Behavioral Detection Fails Where Models Are Wrong The Impact of Off-Policy Training Data on Probe Generalisation
Reference 36
- Resolution
- unresolved
- no resolver link, observed 2026-07-30T21:44:25.201514Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.