Paper Citation Record · LEDGER

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models

As of 21 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 0 inbound Pith citation observations for arXiv:2505.07846.

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

pith.paper-citation-record.v1
2505.07846 v1

Coverage vector

measured 20 of 20 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-15T23:38:47.719414Z

measured 20 of 20 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-21T06:32:19.484+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

20 of 20 outbound references displayed

verified exact0
verified fuzzy6
unresolved14
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 74a4ee3d-e3cd-4bc8-b2ff-f5fdf2502064 · outbound

This paper cites Concrete Problems in AI Safety.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Concrete Problems in AI Safety

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.644757Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce40c890-fc86-49da-a3e3-75833add6ba5 · outbound

This paper cites arXiv preprint (2024).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models arXiv preprint (2024)

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.962400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation df69f904-beb4-4106-97af-4672cf51928f · outbound

This paper cites an unresolved cited work.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Unresolved cited work

Reference 3

Resolution: unresolved; raw_fallback, observed 2026-08-15T23:38:47.951904Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3d02fa8d-4493-4bea-b744-61adf35b4421 · outbound

This paper cites OpenAI Blog (2016).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models OpenAI Blog (2016)

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.940191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 84f7275d-358f-470e-b425-3c3c03af6f93 · outbound

This paper cites Machine Learning 110(9), 2419–2468 (2021).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Machine Learning 110(9), 2419–2468 (2021)

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.928278Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ce182035-b579-4b32-b604-81def3c69671 · outbound

This paper cites AGI Safety Literature Review.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models AGI Safety Literature Review

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.663206Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d2ba8f9-034e-40a1-9963-32fff295183b · outbound

This paper cites Alignment faking in large language models.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Alignment faking in large language models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.667557Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 827234f4-52fb-4446-962d-4aaedcef1b92 · outbound

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

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.671533Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91daad12-abe3-4410-90ed-77a445b796b5 · outbound

This paper cites an unresolved cited work.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Unresolved cited work

Reference 9

Resolution: unresolved; raw_fallback, observed 2026-08-15T23:38:47.916232Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 2bee653f-9818-4781-883d-24c5d9fcc308 · outbound

This paper cites Artificial Life 26(2), 274–306 (2020).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Artificial Life 26(2), 274–306 (2020)

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.905056Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 1b926d21-bbe1-47ce-8be4-18f9098e2038 · outbound

This paper cites Sycophancy in Large Language Models: Causes and Mitigations.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Sycophancy in Large Language Models: Causes and Mitigations

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.682170Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1d1d8f6f-8bec-4ccf-a4f1-f92a939a456c · outbound

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

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Frontier Models are Capable of In-context Scheming

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.687048Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb33a99a-56e9-436b-864c-7b878b2e3aab · outbound

This paper cites METR Technical Report (2024).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models METR Technical Report (2024)

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.893800Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6f2c69fd-02e4-4b75-bfc2-96c83cc33b55 · outbound

This paper cites LLM Agent Honeypot: Monitoring AI Hacking Agents in the Wild.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models LLM Agent Honeypot: Monitoring AI Hacking Agents in the Wild

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.695578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f1b8a99-c9c3-4718-9bd0-cbb3d6ee5358 · outbound

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

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Large Language Models can Strategically Deceive their Users when Put Under Pressure

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.699617Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8fd15b43-0ff9-4c81-b923-a15576c728a8 · outbound

This paper cites Hacking CTFs with Plain Agents.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Hacking CTFs with Plain Agents

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.703644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51597e0b-b3a0-4354-954e-69625c70ee25 · outbound

This paper cites Badllama 3: removing safety finetuning from Llama 3 in minutes.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Badllama 3: removing safety finetuning from Llama 3 in minutes

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.707598Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fcee41c5-caa6-4df1-a157-056f52985aed · outbound

This paper cites AI Sandbagging: Language Models can Strategically Underperform on Evaluations.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models AI Sandbagging: Language Models can Strategically Underperform on Evaluations

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.711493Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5ac26c4b-aba5-44e5-8528-2581966b95c1 · outbound

This paper cites Nature 412(6844), 331–333 (2001).

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models Nature 412(6844), 331–333 (2001)

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T23:38:47.882647Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 9f181e63-96d6-488c-a14a-d9f7f38c2eed · outbound

This paper cites ReAct: Synergizing Reasoning and Acting in Language Models.

Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models ReAct: Synergizing Reasoning and Acting in Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-15T23:38:47.719414Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.