Paper Citation Record · LEDGER

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

As of 21 July 2026, this Paper Citation Record lists 100 of 298 outbound references and 50 inbound Pith citation observations for arXiv:2406.10162.

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

pith.paper-citation-record.v1
2406.10162 v3

Coverage vector

measured 100 of 298 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-17T14:43:29.496457Z

measured 150 of 150 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-07-21T06:31:05.380196+00:00

measured 50 of 50 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-14T13:32:31.523845Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-10T11:57:03.262704Z

Reference resolution

100 of 298 outbound references displayed

verified exact17
verified fuzzy77
unresolved0
parse uncertain0
malformed identifier0
metadata mismatch6

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 8980323e-06f7-4c02-8870-1521160a9b90 · outbound

This paper cites Thinking fast and slow with deep learning and tree search.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Thinking fast and slow with deep learning and tree search

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.899406Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2e9f8a1c-04b1-482c-8616-64badf5023b0 · outbound

This paper cites Understanding strategic deception and deceptive alignment, 9 2023.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Understanding strategic deception and deceptive alignment, 9 2023

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.924808Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 43e57f36-0d1e-4050-b4fe-5d0e1840076f · outbound

This paper cites A general language assistant as a laboratory for alignment.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models A general language assistant as a laboratory for alignment

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.327950Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2627f3db-a064-4d79-991c-bda621d97569 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Constitutional AI: Harmlessness from AI Feedback

Reference 4

Resolution: metadata mismatch; local_arxiv, observed 2026-05-17T14:43:30.051332Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fff303b3-92dd-421d-8a26-dfda33445add · outbound

This paper cites Taken out of context: On measuring situational awareness in llms.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Taken out of context: On measuring situational awareness in llms

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.910997Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 5154eb35-37f0-4c7f-903e-772b14bd91c6 · outbound

This paper cites How useful is quantilization for mitigating specification-gaming? In Safe Machine Learning (SafeML) Workshop at ICLR 2019.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models How useful is quantilization for mitigating specification-gaming? In Safe Machine Learning (SafeML) Workshop at ICLR 2019

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.935435Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a5811194-250f-4d18-9c26-241ecd99179a · outbound

This paper cites Poisoning Web-Scale Training Datasets is Practical.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Poisoning Web-Scale Training Datasets is Practical

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.056914Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 15608572-ea6e-45f9-a237-fbd78b1d951d · outbound

This paper cites Ai in software engineering at google: Progress and the path ahead, June 2024.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Ai in software engineering at google: Progress and the path ahead, June 2024

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.931995Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 7a0a3c17-543f-4944-9f4a-44f4d7aab9da · outbound

This paper cites Faulty reward functions in the wild, 12 2016.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Faulty reward functions in the wild, 12 2016

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.938764Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a4f56780-5e0e-47d6-b1c4-5191d3394511 · outbound

This paper cites Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.945729Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0567688c-87fe-488a-8489-596d3ae72864 · outbound

This paper cites Reward tampering problems and solutions in reinforcement learning: A causal influence diagram perspective.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Reward tampering problems and solutions in reinforcement learning: A causal influence diagram perspective

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.581241Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c6c6de70-1728-436f-bd12-71d56ea03155 · outbound

This paper cites Wichmann.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Wichmann

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.544002Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4fe91c92-6657-4f99-b33a-8119d1834fa9 · outbound

This paper cites Explaining and Harnessing Adversarial Examples.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Explaining and Harnessing Adversarial Examples

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.062305Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 586be625-0b5d-4e3b-8f51-0c5ef947fe5f · outbound

This paper cites Risks from Learned Optimization in Advanced Machine Learning Systems.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Risks from Learned Optimization in Advanced Machine Learning Systems

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.066799Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e9e8c1bd-aa4b-44c2-aea7-3152d8cd212c · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.530948Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c1c93810-b3e6-48fc-bd4f-fd61c207904c · outbound

This paper cites Instrumental deception and manipulation in llms - a case study.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Instrumental deception and manipulation in llms - a case study

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.503629Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e24253c9-5dff-45d8-a0cc-6c2e91e7d805 · outbound

This paper cites User tampering in reinforcement learning recommender systems.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models User tampering in reinforcement learning recommender systems

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.485194Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 172f0c4f-f4f9-445c-a7bc-e6b84da104e0 · outbound

This paper cites Objective robustness in deep reinforcement learning, 05 2021.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Objective robustness in deep reinforcement learning, 05 2021

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.382461Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation cd4ce58b-2c32-4ec6-834f-5853a8b37d18 · outbound

This paper cites Specification gaming: the flip side of ai ingenuity, April 2020.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Specification gaming: the flip side of ai ingenuity, April 2020

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.358723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation d7991abd-adc8-4f5d-9299-b198dd8924d8 · outbound

This paper cites Adversarial examples in the physical world.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Adversarial examples in the physical world

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.003830Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2b8539e9-90c5-42e7-b51b-f6d7b4df4366 · outbound

This paper cites Essai philosophique sur les probabilit \'e s.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Essai philosophique sur les probabilit \'e s

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.045888Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 64526217-4713-4084-b97f-80909a1d07e8 · outbound

This paper cites Towards deep learning models resistant to adversarial attacks.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Towards deep learning models resistant to adversarial attacks

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.081368Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 02176171-9bd9-4b60-ba00-eac73535d627 · outbound

This paper cites Ng, Daishi Harada, and Stuart J.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Ng, Daishi Harada, and Stuart J

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.096325Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 27d2ff29-bed3-49c2-90d3-4b73e8e605e5 · outbound

This paper cites Reward hacking behavior can generalize across tasks.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Reward hacking behavior can generalize across tasks

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.141658Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4cf18075-5ecb-42d5-9283-e1aa0cc5fb9e · outbound

This paper cites The effects of reward misspecification: Mapping and mitigating misaligned models.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models The effects of reward misspecification: Mapping and mitigating misaligned models

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.965448Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0622b12a-7864-4d78-9bd0-180feedbee3b · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.039227Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 92aa62a9-5f65-42e4-abf6-7cdba4b6d426 · outbound

This paper cites Bowman, Amanda Askell, Roger Grosse, Danny Hernandez, Deep Ganguli, Evan Hubinger, Nicholas Schiefer, and Jared Kaplan.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Bowman, Amanda Askell, Roger Grosse, Danny Hernandez, Deep Ganguli, Evan Hubinger, Nicholas Schiefer, and Jared Kaplan

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.959177Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fd2d9ded-fd1f-411d-9dba-365ec904edab · outbound

This paper cites Universal jailbreak backdoors from poisoned human feedback.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Universal jailbreak backdoors from poisoned human feedback

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.975482Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 99c5955f-b427-4fc1-a234-608d2afe4e86 · outbound

This paper cites why should i trust you?.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models why should i trust you?

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.109981Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 534f43aa-5d57-4887-93c9-c3b04823db74 · outbound

This paper cites Proximal Policy Optimization Algorithms.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Proximal Policy Optimization Algorithms

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.071035Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation ff08dd2b-4ae2-4158-8f3b-09110c06d068 · outbound

This paper cites Bowman, Newton Cheng, Esin Durmus, Zac Hatfield-Dodds, Scott R.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Bowman, Newton Cheng, Esin Durmus, Zac Hatfield-Dodds, Scott R

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.018863Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b97224a1-8448-477f-8ca7-5663a03c84c9 · outbound

This paper cites On the exploitability of instruction tuning.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models On the exploitability of instruction tuning

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.060940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b6939a4c-e35b-40e3-9500-15cde7aa50e1 · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.949160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation cd6b8afb-32a5-49c0-a760-12aeb3181c58 · outbound

This paper cites Inducing unprompted misalignment in llms.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Inducing unprompted misalignment in llms

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.135639Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c1a21c75-3efd-4efb-b2b7-61fbb17bd6bd · outbound

This paper cites Intriguing properties of neural networks.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Intriguing properties of neural networks

Reference 36

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.076129Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fb9d9e0d-371f-45ac-bf7b-5fbac537c329 · outbound

This paper cites Active learning helps pretrained models learn the intended task.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Active learning helps pretrained models learn the intended task

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.028287Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation ba0a7c1f-bd17-4b2c-876b-4e66dcebbc31 · outbound

This paper cites Avoiding Tampering Incentives in Deep RL via Decoupled Approval.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Avoiding Tampering Incentives in Deep RL via Decoupled Approval

Reference 38

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.081240Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e2bfffde-b59b-4f12-9430-f213802d603b · outbound

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

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

Reference 39

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.086105Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 91e8ccab-0ea9-40a8-897f-25247a23c0d8 · outbound

This paper cites Schmidt, Jan Hendrik Metzen, and J.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Schmidt, Jan Hendrik Metzen, and J

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.077063Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4514e864-a70a-4ecf-a5cc-ccd2ea171d14 · outbound

This paper cites 2018 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2018 , eprint=

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:46.980061Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2f07a20f-46a6-437f-8424-1836db335841 · outbound

This paper cites Wichmann.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Wichmann

Reference 43

Resolution: verified exact; doi, observed 2026-05-17T14:43:29.870043Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 85f9b4c3-ffe7-49fc-b562-9276c2b6a76d · outbound

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

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models International Conference on Learning Representations , year=

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.104884Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0ef41fb2-06b4-4e3f-8c30-0709e66d2120 · outbound

This paper cites 2017 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2017 , eprint=

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.067100Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0333bc31-7844-4e7e-9a0e-e44de01f0676 · outbound

This paper cites 2024 , month=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2024 , month=

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.024917Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation bccb87c5-fbfa-4b82-a301-94e22ea6b12f · outbound

This paper cites Anthropic News , note=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Anthropic News , note=

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.036659Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c6f41973-0ca3-4f73-a03a-fad4be24aead · outbound

This paper cites AI Alignment Forum , note=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models AI Alignment Forum , note=

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.007680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fb982f2a-bc9e-436e-bf7a-7a939d22f6a1 · outbound

This paper cites LessWrong , note=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models LessWrong , note=

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.031123Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation d28a7c06-7be4-4bd7-ba6b-6a479923df32 · outbound

This paper cites AI Alignment Forum , note=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models AI Alignment Forum , note=

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.129407Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e6901b6f-109c-45e4-ba03-3233bb12274c · outbound

This paper cites 2018 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) , pages=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2018 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) , pages=

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.594056Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f03b65ef-85ea-469f-a863-d01c96c259e5 · outbound

This paper cites Proceedings of the Sixteenth International Conference on Machine Learning (ICML 1999) , pages=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Proceedings of the Sixteenth International Conference on Machine Learning (ICML 1999) , pages=

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.587522Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 98e65236-3639-4f1d-8259-30c8d5c7182f · outbound

This paper cites Safe Machine Learning (SafeML) Workshop at ICLR 2019 , year=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Safe Machine Learning (SafeML) Workshop at ICLR 2019 , year=

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.584359Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4001a806-2d1a-4de7-b275-2dae4b673cb1 · outbound

This paper cites 2019 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2019 , eprint=

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.578068Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 6695ce14-c68a-4b43-9018-c001bff64336 · outbound

This paper cites 2017 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2017 , eprint=

Reference 55

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.575070Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation add757e3-9109-4ac6-98d7-d6bbebbb4c61 · outbound

This paper cites BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain

Reference 56

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.091068Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0d3437d1-119c-420f-abe0-1bf9dd322793 · outbound

This paper cites ArXiv , year=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models ArXiv , year=

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.572097Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 3a6039d2-c44f-4296-a36d-a3dc5015d3e2 · outbound

This paper cites Essai philosophique sur les probabilit.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Essai philosophique sur les probabilit

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.515690Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 750bb235-759d-463d-adad-06762749342f · outbound

This paper cites Wilson, J.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Wilson, J

Reference 59

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:29.931227Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4b7ea378-7d53-454f-a6ab-12f481ece0b8 · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.512815Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 4c1ea0db-47f8-4f5c-afec-b97d80dc8049 · outbound

This paper cites Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society , year=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society , year=

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.509988Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 007197ca-90af-493a-aa6a-df912ee4854f · outbound

This paper cites author=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models author=

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.506747Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 3de36de2-174c-427a-b525-63637fbff7cd · outbound

This paper cites Understanding the Failure Modes of Out-of-Distribution Generalization.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Understanding the Failure Modes of Out-of-Distribution Generalization

Reference 63

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.096550Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 11095b84-6ff6-470c-9741-ef0ee83152dd · outbound

This paper cites 2022 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2022 , eprint=

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.496775Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fb6f45c9-46e5-4d75-9c0f-bd85e9d18fe0 · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.475747Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 1dd97463-319e-4d92-aed0-7935143673a1 · outbound

This paper cites 2020 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2020 , eprint=

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.471689Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 35fa0ce5-92aa-4b6d-bfa1-5a8f9ccc9dd9 · outbound

This paper cites Why Should I Trust You?.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Why Should I Trust You?

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.467652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 9982bc49-ad63-499a-b889-dccce281d6d0 · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.451432Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b3b42f32-197c-4408-9aa7-5298e5d08f40 · outbound

This paper cites 2024 , url=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2024 , url=

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.438194Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f4e7a1b2-4dc9-4125-a7fc-a49423ea369c · outbound

This paper cites 2022 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2022 , eprint=

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.435329Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 46cdaa1c-f1c1-4134-80f2-54f8a838ac16 · outbound

This paper cites 2016 , month=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2016 , month=

Reference 71

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.432263Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 3a3b898b-98b1-41f4-a3e2-77c37f64f53b · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.429404Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 022246db-d4b2-40ed-b4cc-125e380c1ce2 · outbound

This paper cites Information Systems Research , volume=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Information Systems Research , volume=

Reference 73

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:29.992996Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 66d111c3-8d98-4ab4-9fa3-e6a0f0d1f8ca · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.426514Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 73201ba7-ead1-4974-855d-9fc271d46dd4 · outbound

This paper cites DeepMind Blog , year =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models DeepMind Blog , year =

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.423400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 1d977e16-3bbb-4d33-9ea7-0c23fb7bef69 · outbound

This paper cites 2021 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2021 , eprint=

Reference 76

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.414247Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 79e04df4-6fb9-4f9f-853b-6756c1c840d5 · outbound

This paper cites 2022 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2022 , eprint=

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.410877Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 76b54a7f-eecf-41fd-8554-fa0e309e406a · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.407354Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 1c3f1d94-7e67-4d7e-9f02-2332a762fadf · outbound

This paper cites 2024 , month =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2024 , month =

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.350831Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a49423ff-e82f-4f2f-bd68-785c2475144c · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.342779Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 658490a3-96dc-4980-9cd6-1b4e6bb50420 · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 81

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.339303Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fd1813cb-0920-4d9a-9275-a4293a3220ac · outbound

This paper cites Goodfellow, H.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Goodfellow, H

Reference 82

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.032628Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 97feda1f-8d90-4169-a965-f519af819aa3 · outbound

This paper cites Gender Bias in Coreference Resolution , booktitle =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Gender Bias in Coreference Resolution , booktitle =

Reference 83

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.331437Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 716b6b5c-07a4-47ae-b365-2fd57be293e8 · outbound

This paper cites 2019 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2019 , eprint=

Reference 84

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.320117Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 40ced69d-bafe-44b7-81dd-a93a04ee343e · outbound

This paper cites 2024 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2024 , eprint=

Reference 85

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.316372Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a27cfbba-5d91-4c8d-85c2-a749af98bc85 · outbound

This paper cites Concrete Problems in AI Safety.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Concrete Problems in AI Safety

Reference 86

Resolution: verified exact; local_arxiv, observed 2026-05-17T14:43:30.117305Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f65c0d68-3228-4f04-9e3a-b4b59c7940ad · outbound

This paper cites 2023 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , eprint=

Reference 87

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.312355Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 8f5d63c2-d2d4-4713-9137-cedb6a3238b5 · outbound

This paper cites Proceedings of the AAAI Conference on Artificial Intelligence , author=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Proceedings of the AAAI Conference on Artificial Intelligence , author=

Reference 88

Resolution: verified exact; doi, observed 2026-05-17T14:43:29.670120Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 6f5ec12d-6b8f-442c-aa91-78aa27b0930f · outbound

This paper cites an unresolved cited work.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Unresolved cited work

Reference 89

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:43:30.308364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 13f8ff31-53ed-4072-a4c1-5930aa349cfe · outbound

This paper cites Margin-based Parallel Corpus Mining with Multilingual Sentence Embeddings.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Margin-based Parallel Corpus Mining with Multilingual Sentence Embeddings

Reference 90

Resolution: verified exact; doi, observed 2026-05-17T14:43:29.683217Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 27859ef5-948e-4029-a273-ed78e453dd10 · outbound

This paper cites 2021 , eprint=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2021 , eprint=

Reference 91

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.118628Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 5d953f0a-bea8-4726-a0c1-38aa5659eacd · outbound

This paper cites Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifiers , volume =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifiers , volume =

Reference 92

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.074151Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e6f30c89-5d7c-4e69-930d-515e40dd3562 · outbound

This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Reference 93

Resolution: metadata mismatch; local_arxiv, observed 2026-05-17T14:43:29.700654Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 7a23dcdd-94e8-4e23-8873-5215cff1715e · outbound

This paper cites Transactions of the Association for Computational Linguistics , volume =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Transactions of the Association for Computational Linguistics , volume =

Reference 94

Resolution: verified exact; doi, observed 2026-05-17T14:43:29.705364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 258c51e0-0cde-410c-856d-fbd2bea53538 · outbound

This paper cites Large Language Models can Strategically Deceive their Users when Put Under Pressure.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Large Language Models can Strategically Deceive their Users when Put Under Pressure

Reference 95

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.125082Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 746e3596-308d-4da8-b88f-f260e5b181b7 · outbound

This paper cites Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation

Reference 96

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.130930Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation d49c51d2-f4b5-4647-938d-f2279e7fc3d1 · outbound

This paper cites Models in the Loop: Aiding Crowdworkers with Generative Annotation Assistants.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Models in the Loop: Aiding Crowdworkers with Generative Annotation Assistants

Reference 97

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.137251Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 261b3510-287d-448e-b895-d71c5c3af60b · outbound

This paper cites Findings of EMNLP , year=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Findings of EMNLP , year=

Reference 98

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.033860Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 64af8f30-a7ac-403f-9038-881daae39404 · outbound

This paper cites Universal Adversarial Attacks on Text Classifiers , year=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Universal Adversarial Attacks on Text Classifiers , year=

Reference 99

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.048928Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 7d24a82d-cf14-439a-8c06-a0ecabaeb58e · outbound

This paper cites NeurIPS 2022 Competition Track , pages=.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models NeurIPS 2022 Competition Track , pages=

Reference 100

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.058201Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 57f2d55c-d6e1-482e-80c3-7237604393b4 · outbound

This paper cites 2023 , author =.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models 2023 , author =

Reference 101

Resolution: verified fuzzy; raw_fallback, observed 2026-05-17T14:44:47.051767Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 785456e1-bea7-41da-9a24-bc6ac472f27c · outbound

This paper cites Red Teaming Deep Neural Networks with Feature Synthesis Tools.

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models Red Teaming Deep Neural Networks with Feature Synthesis Tools

Reference 102

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.144183Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Pith citing papers

Observation 4fb0d664-6f8e-4828-b17d-c8e4117718cf · inbound

Frontier Models are Capable of In-context Scheming cites this paper.

Frontier Models are Capable of In-context Scheming Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a5a05f9c-4ace-44cb-a4a4-b9b6b38c26a3 · inbound

Alignment faking in large language models cites this paper.

Alignment faking in large language models Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e2a01161-91f0-487b-a661-a31dc246608b · inbound

AI Realtor: Towards Grounded Persuasive Language Generation for Automated Copywriting cites this paper.

AI Realtor: Towards Grounded Persuasive Language Generation for Automated Copywriting Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-05-23T02:57:26.362906Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 7959c5d5-1e1f-4619-8c16-28e31cc59b9b · inbound

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

Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 74

Resolution: metadata mismatch; local_arxiv, observed 2026-05-21T07:24:13.021176Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation e2d588e2-fb9b-4a59-a291-399763e5552d · inbound

VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model cites this paper.

VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation d2d6bf6f-e464-481f-a3b1-947f54e0d5c6 · inbound

Scheming Ability in LLM-to-LLM Strategic Interactions cites this paper.

Scheming Ability in LLM-to-LLM Strategic Interactions Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-05-18T07:51:03.880478Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c8b301e5-4b4b-48bd-9565-55fb4453a628 · inbound

User Detection and Response Patterns of Sycophantic Behavior in Conversational AI cites this paper.

User Detection and Response Patterns of Sycophantic Behavior in Conversational AI Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 850290f5-1f27-4c80-ba82-57fe7263e17f · inbound

Interpretable Electrophysiological Features of Resting-State EEG Capture Cortical Network Dynamics in Parkinsons Disease cites this paper.

Interpretable Electrophysiological Features of Resting-State EEG Capture Cortical Network Dynamics in Parkinsons Disease Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 6

Resolution: unresolved; no resolver link, observed 2026-07-13T14:23:49.346727Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6453c93-e038-4e57-a0ee-1a5e1cdffb0c · inbound

Mitigating LLM biases toward spurious social contexts using direct preference optimization cites this paper.

Mitigating LLM biases toward spurious social contexts using direct preference optimization Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 5

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 503dde2b-ad1c-4039-80a4-32d0c4bf66d3 · inbound

Beyond Semantic Manipulation: Token-Space Attacks on Reward Models cites this paper.

Beyond Semantic Manipulation: Token-Space Attacks on Reward Models Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation cccddd86-e0d9-4867-8fc6-e895e98a8613 · inbound

Can Coding Agents Be General Agents? cites this paper.

Can Coding Agents Be General Agents? Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 15

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation d1174c00-9257-4c9a-b3d4-42fbe2475d1b · inbound

Beyond Static Snapshots: A Grounded Evaluation Framework for Language Models at the Agentic Frontier cites this paper.

Beyond Static Snapshots: A Grounded Evaluation Framework for Language Models at the Agentic Frontier Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0c55d3be-af72-40f4-968a-f8bad6eac925 · inbound

Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories cites this paper.

Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b28d615c-c04b-4e52-a329-068aa41f8631 · inbound

Measuring Opinion Bias and Sycophancy via LLM-based Persuasion cites this paper.

Measuring Opinion Bias and Sycophancy via LLM-based Persuasion Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 3

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c4b3d4fd-a4ff-47e1-9651-be7702087f06 · inbound

Do Prompt-Elicited Trajectories Reflect Training-Time Reward Hacking? A Systematic Study on Monitoring Trainig-Time Reward Hacking in Code Generation cites this paper.

Do Prompt-Elicited Trajectories Reflect Training-Time Reward Hacking? A Systematic Study on Monitoring Trainig-Time Reward Hacking in Code Generation Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation becc456a-2e16-4f32-a635-599da5155290 · inbound

Do Prompt-Elicited Trajectories Reflect Training-Time Reward Hacking? A Systematic Study on Monitoring Trainig-Time Reward Hacking in Code Generation cites this paper.

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-07-01T10:05:40.171174Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b4af3104-1a7f-4deb-9d2f-1669321c1666 · inbound

What Makes a Good Terminal-Agent Benchmark Task: A Guideline for Adversarial, Difficult, and Legible Evaluation Design cites this paper.

What Makes a Good Terminal-Agent Benchmark Task: A Guideline for Adversarial, Difficult, and Legible Evaluation Design Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation ad3f85e8-1cb4-4efe-ab5c-ef8105445446 · inbound

Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity cites this paper.

Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 23

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 40a635cc-cd21-44ed-be2e-be92d47be9e5 · inbound

Explanation Fairness in Large Language Models: An Empirical Analysis of Disparities in How LLMs Justify Decisions Across Demographic Groups cites this paper.

Explanation Fairness in Large Language Models: An Empirical Analysis of Disparities in How LLMs Justify Decisions Across Demographic Groups Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation dcf2e410-a702-4925-9ad1-6e2cc7f80457 · inbound

Do Androids Dream of Breaking the Game? Systematically Auditing AI Agent Benchmarks with BenchJack cites this paper.

Do Androids Dream of Breaking the Game? Systematically Auditing AI Agent Benchmarks with BenchJack Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 16

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2fd81e13-c1a7-4001-b813-c3dc5f9eedf5 · inbound

LLM-Based Persuasion Enables Guardrail Override in Frontier LLMs cites this paper.

LLM-Based Persuasion Enables Guardrail Override in Frontier LLMs Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 16

Resolution: metadata mismatch; arxiv_id, observed 2026-05-17T14:43:30.598089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 9be1ab62-55b2-4938-9447-a40d42bd83ce · inbound

Enhancing the Code Reasoning Capabilities of LLMs via Consistency-based Reinforcement Learning cites this paper.

Enhancing the Code Reasoning Capabilities of LLMs via Consistency-based Reinforcement Learning Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 7

Resolution: metadata mismatch; local_arxiv, observed 2026-05-20T13:18:18.496423Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 5e2c20d4-88f5-4113-9cec-c6e4d41aed68 · inbound

Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale cites this paper.

Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 25

Resolution: verified exact; local_arxiv, observed 2026-05-21T06:59:45.584507Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 163fa1bc-2ac2-40b9-86f8-85da06afb0f4 · inbound

SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents cites this paper.

SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-21T03:09:28.096652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 43c941bf-8398-46df-95ab-a3bb1140c49f · inbound

Cultivating Machine Intelligence: The OMEGA Shift from Top-Down Optimization to Autopoietic Cognitive Ecologies cites this paper.

Cultivating Machine Intelligence: The OMEGA Shift from Top-Down Optimization to Autopoietic Cognitive Ecologies Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-06-30T00:04:06.708859Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 8e1b49bd-7dde-412f-a288-9dd24e45a4cc · inbound

Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm cites this paper.

Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 35

Resolution: verified exact; local_arxiv, observed 2026-06-29T13:03:26.784254Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f5c5f782-ca2d-4334-83a5-bbdf8fcaa1fe · inbound

VeriGate: Verifier-Gated Step-Level Supervision for GRPO cites this paper.

VeriGate: Verifier-Gated Step-Level Supervision for GRPO Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-06-29T08:53:16.023789Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 9241f516-fe1e-4869-b5f7-959bdb85e8c7 · inbound

FORGE: Multi-Agent Graduated Exploitation and Detection Engineering cites this paper.

FORGE: Multi-Agent Graduated Exploitation and Detection Engineering Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 7

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T03:46:32.324436Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fc16fbca-74f8-41dc-90dd-e3b490d9c523 · inbound

Large Language Models Hack Rewards, and Society cites this paper.

Large Language Models Hack Rewards, and Society Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 53

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T01:46:26.998730Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fb42a312-2899-4dcb-b526-a34d427c02b4 · inbound

Coding with "Enemy": Can Human Developers Detect AI Agent Sabotage? cites this paper.

Coding with "Enemy": Can Human Developers Detect AI Agent Sabotage? Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 39

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T12:46:57.256904Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 11ccc3bd-39ea-47f6-866b-26bb6a4572af · inbound

CogManip: Benchmarking Manipulative Behavior in Multi-Turn Interactions with Large Language Model cites this paper.

CogManip: Benchmarking Manipulative Behavior in Multi-Turn Interactions with Large Language Model Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 12

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T12:56:57.391611Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 394db164-97a6-452b-af4f-66a30eff9a4e · inbound

Position: Anthropomorphic Misalignment Research Needs Stronger Evidence cites this paper.

Position: Anthropomorphic Misalignment Research Needs Stronger Evidence Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 160

Resolution: metadata mismatch; local_arxiv, observed 2026-06-28T20:22:37.288358Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 021bc0e7-47b4-4465-adbf-366d854ed7eb · inbound

Building Comparative Motivation Profiles with Instrumental Interventions cites this paper.

Building Comparative Motivation Profiles with Instrumental Interventions Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 15

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T21:27:24.578512Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c3cb19e1-bb26-4255-9b75-c2ec21ae5fe9 · inbound

Projecting the Emerging Mindset of SWE Agent by Launching a Wild Code Understanding Journey cites this paper.

Projecting the Emerging Mindset of SWE Agent by Launching a Wild Code Understanding Journey Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 20

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T23:17:29.846239Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation afbb7d4b-b34e-4d38-bcbb-c960a5214419 · inbound

Activation Steering Induces Emergent Misalignment: A More Comprehensive Evaluation cites this paper.

Activation Steering Induces Emergent Misalignment: A More Comprehensive Evaluation Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 4

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T22:57:26.132085Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 13334f28-f0b2-45fb-8562-6b4f55bf5f4c · inbound

The Hidden Bias of Process Reward Models:PRISM for Rewarding the Right Reasoning cites this paper.

The Hidden Bias of Process Reward Models:PRISM for Rewarding the Right Reasoning Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 6

Resolution: metadata mismatch; local_arxiv, observed 2026-07-03T00:37:30.088793Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation abec1965-37cc-47c3-9e30-c79f7d008d73 · inbound

Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization cites this paper.

Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 10

Resolution: metadata mismatch; local_arxiv, observed 2026-07-03T01:37:30.246728Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b048d25a-165b-4535-94dd-fb1c50cdf804 · inbound

Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization cites this paper.

Reference 122

Resolution: metadata mismatch; local_arxiv, observed 2026-06-27T16:31:02.711701Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c2c97127-db66-4e72-a608-f67915f0aac8 · inbound

The Distributed Detectability Band Against Marginal-Preserving Attacks cites this paper.

The Distributed Detectability Band Against Marginal-Preserving Attacks Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-07-03T05:57:41.857798Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 695a078a-1ada-47bd-a064-7bb2cf29d4cd · inbound

AI Coding Agents in Social Science: Methodologically Diverse, Empirically Consistent, Interpretively Vulnerable cites this paper.

AI Coding Agents in Social Science: Methodologically Diverse, Empirically Consistent, Interpretively Vulnerable Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 23

Resolution: verified exact; local_arxiv, observed 2026-07-03T05:47:41.516909Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 1f6040cb-2e5d-4f5f-a0f6-0ef4f06ebe63 · inbound

Helpful or Harmful? Evaluating LLM-Assisted Vulnerability Patching via a Human Study cites this paper.

Helpful or Harmful? Evaluating LLM-Assisted Vulnerability Patching via a Human Study Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-07-04T21:10:09.261675Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c87b7cf3-90a7-4f11-a3e1-a65931e5cb4f · inbound

Reframing AGI Confrontation with Off Earth Autonomy cites this paper.

Reframing AGI Confrontation with Off Earth Autonomy Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 10

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T08:35:34.308556Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 039872d3-ea39-4fc4-8aae-81fc21567d27 · inbound

Evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment cites this paper.

Evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 14

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T09:45:39.620492Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 213daf82-afcc-4978-a8ab-bfddf950c0e9 · inbound

Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting cites this paper.

Stop Hand-Holding Your Coding Agent: Engineering the Loops that Replace Step-by-Step Prompting Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-07-02T20:47:22.229284Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 306e0df1-c2ef-4643-8799-d1f0f98ce1ac · inbound

MemSyco-Bench: Benchmarking Sycophancy in Agent Memory cites this paper.

MemSyco-Bench: Benchmarking Sycophancy in Agent Memory Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 10

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T06:36:43.279118Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 51465a93-53b0-4846-af79-904c1d71c737 · inbound

MemSyco-Bench: Benchmarking Sycophancy in Agent Memory cites this paper.

MemSyco-Bench: Benchmarking Sycophancy in Agent Memory Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 10

Resolution: metadata mismatch; local_arxiv, observed 2026-07-03T18:58:50.747148Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 2cebb91a-2087-4052-a83f-eebe9b816bef · inbound

How to Avoid Debate: Scalable AI Safety via Doubly-Efficient Interactive Proofs cites this paper.

How to Avoid Debate: Scalable AI Safety via Doubly-Efficient Interactive Proofs Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-07-12T01:34:45.323277Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4cfe71e8-1748-473d-8a94-d98327c9602c · inbound

Overthinking: Amplifying Reasoning Weights to Extract Learned Secrets cites this paper.

Overthinking: Amplifying Reasoning Weights to Extract Learned Secrets Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 9

Resolution: metadata mismatch; local_arxiv, observed 2026-07-10T11:57:03.264446Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 9aa4ef7f-60e8-4794-8caa-439a9bb2d657 · inbound

A small language model detects behavioural faithfulness gaps that frontier judges and human raters miss cites this paper.

A small language model detects behavioural faithfulness gaps that frontier judges and human raters miss Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 81

Resolution: unresolved; no resolver link, observed 2026-07-13T04:01:29.587725Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66ee4a79-8638-4e32-9a33-b90c0d16b2e9 · inbound

Two Confounds in Cross-Model Value Comparison: Response Determinism and the Access Harness cites this paper.

Two Confounds in Cross-Model Value Comparison: Response Determinism and the Access Harness Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-14T13:32:31.523845Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.