Paper Citation Record · LEDGER

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction

As of 18 August 2026, this Paper Citation Record lists 43 of 43 outbound references and 2 inbound Pith citation observations for arXiv:2508.04216.

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

pith.paper-citation-record.v1
2508.04216 v1

Coverage vector

measured 43 of 43 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T00:49:58.910725Z

measured 45 of 45 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-06-29T12:22:06.635622Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-06-29T12:23:24.240761Z

Reference resolution

43 of 43 outbound references displayed

verified exact0
verified fuzzy4
unresolved39
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 53475a56-da37-4582-bb18-dfc7b62a9ad8 · outbound

This paper cites , " * write output.state after.block = add.period write newline.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction , " * write output.state after.block = add.period write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.195733Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a21285f0-4606-462a-8b62-e26e1c89cb44 · outbound

This paper cites write newline.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction write newline

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.287688Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7953bb5a-bb0d-4a5f-bbc0-e78ea0e11043 · outbound

This paper cites Concrete Problems in AI Safety.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Concrete Problems in AI Safety

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.417778Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 426f3b28-a921-476d-88da-4a3fe872c68c · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 4

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:02.140984Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 870055df-2744-4273-8716-6b2db294e08f · outbound

This paper cites E.; Hume, T.; Carter, S.; Henighan, T.; and Olah, C.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction E.; Hume, T.; Carter, S.; Henighan, T.; and Olah, C

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.665028Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d56d4c56-8dea-4b89-a9fe-5ff1ff83ac11 · outbound

This paper cites Large Language Monkeys: Scaling Inference Compute with Repeated Sampling.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.743365Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7532f3ab-76f8-46a2-9748-bac3186db987 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Training Verifiers to Solve Math Word Problems

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.811698Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4fed5e4-9b22-432b-9197-85b9a818c276 · outbound

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

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.891955Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 48843abe-3a9f-443d-aca0-0ce19980c0d0 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:55.975299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c175fe98-219e-43b5-be7c-dcaddc4ccf1d · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:01.974682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6b720003-3c15-4315-87ef-c01af2f16297 · outbound

This paper cites The Llama 3 Herd of Models.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction The Llama 3 Herd of Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.114202Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff348be9-27de-427a-8ce6-a0f67c93a558 · outbound

This paper cites Inverse Reward Design.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Inverse Reward Design

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.182077Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 993c3d33-30a8-4051-9fcd-da0bce7e8267 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 13

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:01.825837Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 02c988b7-9dd4-47ae-bba3-607f415acebc · outbound

This paper cites Let's Verify Step by Step.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Let's Verify Step by Step

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.306521Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c1a11d3-de30-4c11-9006-604f787b15e8 · outbound

This paper cites Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.372971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 151a3599-716c-4f98-ad09-f762cebc36b2 · outbound

This paper cites RRM: Robust Reward Model Training Mitigates Reward Hacking.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction RRM: Robust Reward Model Training Mitigates Reward Hacking

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.466791Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2786640-1018-43ad-8f4e-99a900c9a45c · outbound

This paper cites P.; Hermann, K.; Welleck, S.; Yazdanbakhsh, A.; and Clark, P.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction P.; Hermann, K.; Welleck, S.; Yazdanbakhsh, A.; and Clark, P

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T00:50:01.604360Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 41967675-9e22-4ff0-9d31-615883143be6 · outbound

This paper cites Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.667273Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f01d6fc3-fd5d-46be-9266-4d1dc61a95e1 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:01.465055Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5fc8f8d2-3bc5-4065-acc1-1db253b092fd · outbound

This paper cites Feedback Loops With Language Models Drive In-Context Reward Hacking.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Feedback Loops With Language Models Drive In-Context Reward Hacking

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.840261Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a8c35825-02bd-434b-9fd9-eacef2c06611 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:56.900164Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b599c292-0fe5-4857-8b69-41716d6dfa30 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 22

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:01.243287Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 11feddce-4347-4d37-8522-ca4ea2a71324 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:01.071894Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c81a857b-41ce-4035-aa10-4432d0b99b1f · outbound

This paper cites Self-critiquing models for assisting human evaluators.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Self-critiquing models for assisting human evaluators

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.156128Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1aef55fa-a437-473b-bcd4-c93e13044071 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:00.812569Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3abde802-218f-44d7-ba48-e00c9527e919 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 26

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:00.557972Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation d45b03e4-3e88-475d-a03f-6465fd3d4c36 · outbound

This paper cites DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.391495Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ca4aa02-c030-4f7d-9025-fd91360e5b06 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.500199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4bf3754-bb08-44c9-8096-667569423794 · outbound

This paper cites Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.585878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f8048e73-95ee-4f11-8082-d620ed4e4900 · outbound

This paper cites V.; Lee, J.; Xu, K.; and Kumar, A.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction V.; Lee, J.; Xu, K.; and Kumar, A

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T00:50:00.332444Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 94254e3b-f5aa-4a07-87f9-b90e7bc10d02 · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 31

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:50:00.119426Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 06cff769-dda9-49ac-877f-d6f01645f5de · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.868476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea48494e-966a-4450-9252-0561cecf8267 · outbound

This paper cites L.; McDougall, C.; MacDiarmid, M.; Freeman, C.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction L.; McDougall, C.; MacDiarmid, M.; Freeman, C

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:57.951694Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fa39720-4336-4778-996c-69b162fd43dc · outbound

This paper cites Solving math word problems with process- and outcome-based feedback.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Solving math word problems with process- and outcome-based feedback

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.030203Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20e4ba43-a24d-4141-b814-b64088f259ae · outbound

This paper cites Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.128960Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5210ed7d-8a0c-4e7d-8e2e-8a4285517db3 · outbound

This paper cites V.; and Zhou, D.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction V.; and Zhou, D

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T00:49:59.909412Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation bac4398b-b0d7-4bdf-9178-1545702fd1ef · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 37

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:49:59.733765Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a68159bb-9a83-431f-834f-94b0921806ab · outbound

This paper cites an unresolved cited work.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Unresolved cited work

Reference 38

Resolution: unresolved; raw_fallback, observed 2026-08-06T00:49:59.525941Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 99f1c65a-14e5-4d15-8569-c13c054177e4 · outbound

This paper cites Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.564042Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 941c3b9b-e322-4766-ac12-5d0c6c6af4b9 · outbound

This paper cites L.; Cao, Y.; and Narasimhan, K.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction L.; Cao, Y.; and Narasimhan, K

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T00:49:59.304045Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 87d5d540-9d51-4cb6-85c3-5b5292754f9a · outbound

This paper cites Automatic Chain of Thought Prompting in Large Language Models.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction Automatic Chain of Thought Prompting in Large Language Models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.718193Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7bb593af-902e-48f0-b2d3-c34250c53f0b · outbound

This paper cites The Lessons of Developing Process Reward Models in Mathematical Reasoning.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction The Lessons of Developing Process Reward Models in Mathematical Reasoning

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.821187Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23122d00-eb26-4d5e-92d0-df77482177dd · outbound

This paper cites A Survey of Large Language Models.

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction A Survey of Large Language Models

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-06T00:49:58.910725Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 74308e21-8960-496a-9d51-81d753f7543e · inbound

Factored Causal Representation Learning for Robust Reward Modeling in RLHF cites this paper.

Factored Causal Representation Learning for Robust Reward Modeling in RLHF Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-21T14:20:13.564528Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 90bf14c4-aaac-4d72-a5a3-94d7926aa0ef · inbound

Reward Bias Substitution: Single-Axis Bias Mitigations Redirect Optimization Pressure cites this paper.

Reward Bias Substitution: Single-Axis Bias Mitigations Redirect Optimization Pressure Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction

Reference 78

Resolution: verified exact; arxiv_id, observed 2026-06-29T12:23:24.242250Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.