Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization | Paper Citation Record

Observation 3dd74b66-7ef7-4216-9965-3c3ee9b83011 · outbound

This paper cites This implies that240k should be in the form ofm3 for some integerm.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization This implies that240k should be in the form ofm3 for some integerm

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.566895Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 1e2f23df-1cb7-4085-bcd7-841fd5c2f693 · outbound

This paper cites Rejected(Unalinged GPT-4) 1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Rejected(Unalinged GPT-4) 1

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.994342Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 34f32bd1-3029-449a-bce9-36d47b9bf3dd · outbound

This paper cites DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.693379Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df890ed8-4339-4fc9-bcc3-c743a8b679ab · outbound

This paper cites RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.698392Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 154d232b-4591-4a66-825d-5bc2c356a39e · outbound

This paper cites Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.703118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 551757cb-7d42-4d87-9df2-a53eae1d4dcc · outbound

This paper cites Advancing LLM Reasoning Generalists with Preference Trees.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Advancing LLM Reasoning Generalists with Preference Trees

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.707744Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f376e46e-84f6-40f2-9f86-c4455c1e8f43 · outbound

This paper cites Secrets of RLHF in Large Language Models Part I: PPO.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Secrets of RLHF in Large Language Models Part I: PPO

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.712424Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b4755f8-f836-45ba-a6ce-69d439f43737 · outbound

This paper cites Table 6: An example of chosen and rejected solution from RewardBench.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Table 6: An example of chosen and rejected solution from RewardBench

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.904089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 505c06a6-df15-4fdf-85e6-c1edca76788f · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 9

Resolution: malformed identifier; no resolver link, observed 2026-08-15T20:34:32.721081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e2e51d5-d15b-49c5-8c03-27c148e7872a · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 11

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.982224Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5a24994f-0bb7-4761-9b51-7bb3c5fe96d7 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 12

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.970124Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 2629dcc6-529e-4d6e-9572-eef3a6809f1e · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 13

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.957557Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation d94c65db-2420-4d5d-87de-18cb3936d8c0 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.943575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation b907e826-497f-4ef8-bd13-2b8d8327d812 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.932442Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 24d136fd-2521-4b43-a4c0-1143a379696d · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.920457Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f735ec23-2f34-4a8d-aa21-9c0be9615ba4 · outbound

This paper cites So, 240 = 24 ×31 ×51.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization So, 240 = 24 ×31 ×51

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.888723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 855849ce-812e-4cf3-8aa0-8e723ffbd518 · outbound

This paper cites For31, add 2 (1 + 2 = 3).

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization For31, add 2 (1 + 2 = 3)

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.874841Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation fc9b169c-68ef-4c55-bfed-6b2e908e2f60 · outbound

This paper cites Chosen Detailed 1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Chosen Detailed 1

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.861146Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation de2ccbe1-f7b1-4e80-b991-2c04a01643fe · outbound

This paper cites We do this by progressively dividing by the smallest prime numbers.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization We do this by progressively dividing by the smallest prime numbers

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.849579Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6f970ad8-fe7d-407a-a238-b761a00a16ee · outbound

This paper cites 120 is even, so divide by 2:120÷2 = 60.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization 120 is even, so divide by 2:120÷2 = 60

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.838370Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ea9fa6f1-ffa3-4d88-96f5-7b4f0bb4a3f8 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.825971Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3b64f641-fcf1-4b97-92e5-d5a1aaede84a · outbound

This paper cites Here, a = 2 (since 4 + 2 = 6).

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Here, a = 2 (since 4 + 2 = 6)

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.814134Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation c4d75061-7ad2-48cb-8054-739c4c7d93fc · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.801610Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 2b795c59-fa15-4a91-ab2f-01c7c1277ef9 · outbound

This paper cites Thus,k = 4×9×25.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Thus,k = 4×9×25

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.789886Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4912dc05-fe63-4f88-8f0e-31229895b0af · outbound

This paper cites Then,36×25 = 900.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Then,36×25 = 900

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.777421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7847273f-5578-4459-bca7-284d422d287d · outbound

This paper cites To solve the problem, we need to find the smallest positive integerk such that the product 240k is a perfect cube.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization To solve the problem, we need to find the smallest positive integerk such that the product 240k is a perfect cube

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.765399Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 75a3dfad-5f39-4cfd-a7b0-a643be3aa5e6 · outbound

This paper cites We do this by progressively dividing by the smallest prime numbers.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization We do this by progressively dividing by the smallest prime numbers

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.753614Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 31d78d8f-55c3-4979-ac13-cba078e7aec4 · outbound

This paper cites - For24, we need4 +a ≡0 (mod 3).

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - For24, we need4 +a ≡0 (mod 3)

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.741544Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 1e25aef0-1c1f-46c0-b7ec-38956c57f036 · outbound

This paper cites Problem The product of integers 240 andk is a perfect cube.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Problem The product of integers 240 andk is a perfect cube

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.729112Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5f52b9ae-297a-47ff-b7cf-03f622e81c81 · outbound

This paper cites -51 needs two more factors of5(52) to reach53.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization -51 needs two more factors of5(52) to reach53

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.714769Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ae7be370-cd0d-4c2b-96f0-6945423a9e95 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 33

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.701140Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 99191700-7604-4364-9a83-ab726bd6a51f · outbound

This paper cites Rejected Detailed 1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Rejected Detailed 1

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.688399Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 80fca837-f18f-4228-a09c-3bdd673aa0cb · outbound

This paper cites Each prime power in the factorization of a number that forms a perfect cube should be a multiple of 3.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Each prime power in the factorization of a number that forms a perfect cube should be a multiple of 3

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.675697Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0e7054af-0219-4777-9c33-38c5c94ec6b7 · outbound

This paper cites For24, we need at least25 to have a power that is a multiple of 3.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization For24, we need at least25 to have a power that is a multiple of 3

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.663641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 84abcaca-dc95-4aaf-86f7-2a44634b76ec · outbound

This paper cites Calculate each component:21 = 2,32 = 9,52 = 25.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Calculate each component:21 = 2,32 = 9,52 = 25

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.651757Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 9d3876c3-af4f-4d1f-9446-2ad716c05573 · outbound

This paper cites Markdown Format 1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Markdown Format 1

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.638195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6e09b614-765c-4a2e-a9bf-b124d98b34ec · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 39

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.625716Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 10994be8-6693-4517-b3cb-41195dced9ab · outbound

This paper cites - For24, we need at least25 to have a power that is a multiple of 3.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - For24, we need at least25 to have a power that is a multiple of 3

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.612811Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation dbbac7d5-29c2-4176-8260-1f103a2e06d7 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 41

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.601064Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 23d50643-8f09-4019-a96a-aadec19cf07f · outbound

This paper cites Problem The product of integers 240 andk is a perfect cube.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Problem The product of integers 240 andk is a perfect cube

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.590047Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5ff82e89-91af-42e8-b0d7-2bc5c789dbea · outbound

This paper cites GPT-3.5-turbo- 0125.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization GPT-3.5-turbo- 0125

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.578491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5f1b8d29-e7ac-44d2-9284-fa65c093df67 · outbound

This paper cites To make240k a perfect cube, we need to add multiples of 2, 3, and 5 to make the powers of all prime factors multiples of 3.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization To make240k a perfect cube, we need to add multiples of 2, 3, and 5 to make the powers of all prime factors multiples of 3

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.554566Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4d997e61-e49d-4072-a64c-52b7933b1309 · outbound

This paper cites Chosen(Random) Llama3-70B- Instruct.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Chosen(Random) Llama3-70B- Instruct

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.542122Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4bd195e9-77b5-4eef-9332-8263de86b118 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 47

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.529873Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8b2e8ed3-a837-4499-ae81-b63a37a81711 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 48

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.518831Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 391bed6f-1f80-40c5-af4b-60fac25cd7db · outbound

This paper cites 4.Therefore, the smallest possible positive value ofk is 900.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization 4.Therefore, the smallest possible positive value ofk is 900

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.506949Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation de89d9d5-268f-4704-9dee-9f070e3468b1 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 50

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.494883Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation cddb7c31-da16-497a-b6cb-49aeec47f5e7 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 51

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.482096Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation fb113322-4031-4806-bb5e-baca09bc4945 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 52

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.470101Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation caaab7bb-ad4a-42f6-a51b-de73af03b9ff · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 53

Resolution: parse uncertain; raw_fallback, observed 2026-08-15T20:34:33.458329Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f41d9989-5ce4-4bd4-ad1e-a2f337ad80c9 · outbound

This paper cites GPT-3.5-turbo- 0125.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization GPT-3.5-turbo- 0125

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.446079Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6c455fbb-dbc3-4411-a169-3910a9677535 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 55

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.435644Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8cbff164-8eb0-4d92-9627-080090220899 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 56

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.423898Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 31faad9f-7f9b-49dc-890c-a33b4411652a · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 57

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.411527Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 126a26e4-9f3a-4046-90a6-d31edcf8682b · outbound

This paper cites Rejected(Random) Gemma2-27B-it 1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Rejected(Random) Gemma2-27B-it 1

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.399390Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation b5f1df76-756c-43fd-93c6-1cc808406dbc · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 59

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.386522Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6fb6afe6-4da2-4e71-bc0d-88a3dded69b1 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 60

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.373945Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 69560c16-f8b8-49da-8fe5-45c43ac47b66 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 61

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.362211Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 206ecfbb-ed79-4635-ba48-0ff05e28bca1 · outbound

This paper cites WizardMath-7B- v1.1.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization WizardMath-7B- v1.1

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.350361Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4720a234-aa17-4da8-a114-9f83e9af44e9 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 63

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.338614Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ee2c1def-7bbd-437a-a671-07bb4c30ce04 · outbound

This paper cites First, we need to find the prime factorization of 240.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization First, we need to find the prime factorization of 240

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.327564Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation da0ca5b2-0dcf-4156-a171-5713dae68018 · outbound

This paper cites A perfect cube is a number that can be expressed as the cube of an integer.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization A perfect cube is a number that can be expressed as the cube of an integer

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.316094Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f3860637-8c7e-45bd-8ac2-d152c343f50a · outbound

This paper cites We need to find the smallestk such that the product of 240 andk is a perfect cube.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization We need to find the smallestk such that the product of 240 andk is a perfect cube

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.304030Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 63417d51-e455-4bb3-8782-69dc2ad548a0 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 67

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.291832Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation df20eb16-ad6d-471f-835d-82f63c05d870 · outbound

This paper cites The smallest possible values fora,b, andc that make3a +b +c divisible by 3 area = 0,b = 1, andc = 0.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization The smallest possible values fora,b, andc that make3a +b +c divisible by 3 area = 0,b = 1, andc = 0

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.281117Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5604e2a9-70d6-40e1-9206-98680906ae84 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 69

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.269971Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f7dfcf50-6a95-4b0c-8b57-b7b74a19b52d · outbound

This paper cites Table 9: An example of chosen (GPT-4* and Random, 3) and rejected (Random, 3) solution.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Table 9: An example of chosen (GPT-4* and Random, 3) and rejected (Random, 3) solution

Reference 70

Resolution: malformed identifier; raw_fallback, observed 2026-08-15T20:34:33.257901Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 79d6aaae-668a-460b-9d19-bf2b57dff193 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 71

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.246855Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 94b05ac8-9864-4a51-b11c-54eb7f0780c2 · outbound

This paper cites overrefusal.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization overrefusal

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.236642Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 447548d9-cb33-464a-9219-90a04e58592d · outbound

This paper cites - There are no unnecessary disclaimers or limitations.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - There are no unnecessary disclaimers or limitations

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.226022Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3c0d095a-885f-4def-9ac8-0ce67e76b543 · outbound

This paper cites - Some answer is provided, yet it is needlessly restricted or accompanied by refusal-like statements.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - Some answer is provided, yet it is needlessly restricted or accompanied by refusal-like statements

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.214589Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6fdf8bca-12bd-42ec-9c13-e9e06e893c50 · outbound

This paper cites - The refusal is not aligned with policy since the prompt is actually allowed.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - The refusal is not aligned with policy since the prompt is actually allowed

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.201653Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0369e169-7881-4b11-b0ff-f1f1d779f04f · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 76

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.188623Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8f1f476a-48c4-462d-a63c-1d600bbffd93 · outbound

This paper cites an unresolved cited work.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Unresolved cited work

Reference 77

Resolution: unresolved; raw_fallback, observed 2026-08-15T20:34:33.176364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 49ac354b-0334-407b-bc20-488365de4c3f · outbound

This paper cites Your task is to determine how the response handles the harmful prompt.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Your task is to determine how the response handles the harmful prompt

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.163252Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 99194f9a-9080-4553-982a-6f49fadbc8b4 · outbound

This paper cites - There is no language indicating refusal.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - There is no language indicating refusal

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.151992Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 378be4bb-433c-407e-89c2-74b42b6e527b · outbound

This paper cites - However, it may be incomplete or include disclaimers, indicating it is not fully complying.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - However, it may be incomplete or include disclaimers, indicating it is not fully complying

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.141121Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 996f3d35-73a0-4055-a4d3-865200753fa0 · outbound

This paper cites - It does not fully comply by giving the requested harmful information.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - It does not fully comply by giving the requested harmful information

Reference 81

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.129658Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 68a8f1e9-f7f8-4720-bb5d-ea0ebf970520 · outbound

This paper cites - It does not provide information or instructions that fulfill the harmful request.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization - It does not provide information or instructions that fulfill the harmful request

Reference 82

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T20:34:33.117867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 07251166-3685-4dca-8b10-33a14f639ed7 · outbound

This paper cites In gen- eral, the reward model is obtained by replacing the final output layer of causal language model with a linear head to predict a scalar.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization In gen- eral, the reward model is obtained by replacing the final output layer of causal language model with a linear head to predict a scalar

Reference 1952

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.716697Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 604f7e18-35cc-4fd8-92d9-7c37da7c66a2 · outbound

This paper cites UltraFeedback: Boosting Language Models with Scaled AI Feedback.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization UltraFeedback: Boosting Language Models with Scaled AI Feedback

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.688492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6a8fccf-9412-48af-956e-32eb4e0390cb · outbound

This paper cites Critique-out-Loud Reward Models.

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization Critique-out-Loud Reward Models

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-15T20:34:32.682880Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers