Paper Citation Record · LEDGER

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions

As of 17 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 2 inbound Pith citation observations for arXiv:2508.21188.

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

pith.paper-citation-record.v1
2508.21188 v2

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-05T14:35:52.198507Z

measured 26 of 26 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-03T10:13:15.414661Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-06-30T21:05:04.650707Z

Reference resolution

24 of 24 outbound references displayed

verified exact0
verified fuzzy1
unresolved22
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation d91e600d-092a-47f8-aa07-3c2213d50cbb · outbound

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

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.599957Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a52c702-ba68-4ee9-9539-dc89e4ac3154 · outbound

This paper cites LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.713715Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 920baffd-8102-4ed6-a9a0-9c5665ef8fb3 · outbound

This paper cites BIG-Bench Extra Hard.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions BIG-Bench Extra Hard

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.806216Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e75d04a6-b8f8-4f1a-b647-0042b164180d · outbound

This paper cites SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.883516Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44a3cdc7-cb6d-4f98-8329-29478b16b20e · outbound

This paper cites The Climb Carves Wisdom Deeper Than the Summit: On the Noisy Rewards in Learning to Reason.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions The Climb Carves Wisdom Deeper Than the Summit: On the Noisy Rewards in Learning to Reason

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.973056Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba9d74c3-dfff-40d4-a8b7-f1245ac45eb0 · outbound

This paper cites KOR-Bench: Benchmarking Language Models on Knowledge-Orthogonal Reasoning Tasks.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions KOR-Bench: Benchmarking Language Models on Knowledge-Orthogonal Reasoning Tasks

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.058824Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6f85f85-7ff3-4df7-ac01-01f864a1883f · outbound

This paper cites The Llama 3 Herd of Models.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions The Llama 3 Herd of Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.138696Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 360502bb-0bf8-443c-8b4b-c54c739bc552 · outbound

This paper cites Qwen2.5 Technical Report.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Qwen2.5 Technical Report

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.218027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44f72023-4147-499e-a3ee-f9bc4bef15eb · outbound

This paper cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.290902Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cde19674-1179-4713-ad61-86d99b27d526 · outbound

This paper cites Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.350639Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ee51b7f-ac07-47e9-89b6-74018e1af83f · outbound

This paper cites Reinforcement Learning for Reasoning in Large Language Models with One Training Example.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Reinforcement Learning for Reasoning in Large Language Models with One Training Example

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.606028Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd79800b-fecd-405a-a6fd-aafd6af5ed3f · outbound

This paper cites Reasoning or memorization? unreliable results of reinforcement learning due to data contamination.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Reasoning or memorization? unreliable results of reinforcement learning due to data contamination

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.701515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6928d6d9-b19f-44b0-8891-3e26d347ef82 · outbound

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

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.798748Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 77b662ac-1ccc-4f5c-8e8b-06c47cd8619c · outbound

This paper cites Qwen3 Technical Report.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Qwen3 Technical Report

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.888457Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f521cb35-0606-48b9-995f-748399cf3d8b · outbound

This paper cites Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.980645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 65520dd4-4391-4260-b429-56cd12c0c684 · outbound

This paper cites The format reward is different from that of Shao et al.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions The format reward is different from that of Shao et al

Reference 23

Resolution: malformed identifier; raw_fallback, observed 2026-08-05T14:35:53.101114Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3b774ea2-a172-4f3a-91cc-561c70806613 · outbound

This paper cites an unresolved cited work.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Unresolved cited work

Reference 24

Resolution: unresolved; raw_fallback, observed 2026-08-05T14:35:52.956984Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f0ac64dc-1cd8-42bb-a316-3a8140bbfc4c · outbound

This paper cites Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

Reference 1998

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.402781Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 505fbb72-68cb-49d4-bd23-1aa857ad6973 · outbound

This paper cites TTRL: Test-Time Reinforcement Learning.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions TTRL: Test-Time Reinforcement Learning

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:52.017655Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b25332ec-6417-42d4-a68f-75cfce780dbf · outbound

This paper cites Measuring Mathematical Problem Solving With the MATH Dataset.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Measuring Mathematical Problem Solving With the MATH Dataset

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.660785Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7925ea58-f691-467b-9db3-a47591b3db85 · outbound

This paper cites Kimi k1.5: Scaling Reinforcement Learning with LLMs.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Kimi k1.5: Scaling Reinforcement Learning with LLMs

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:51.511047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 92c72353-49a2-4e9c-9e84-ad284e08276f · outbound

This paper cites Jiangjie Chen, Qianyu He, Siyu Yuan, Aili Chen, Zhicheng Cai, Weinan Dai, Hongli Yu, Qiying Yu, Xuefeng Li, Jiaze Chen, Hao Zhou, and Mingxuan Wang.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Jiangjie Chen, Qianyu He, Siyu Yuan, Aili Chen, Zhicheng Cai, Weinan Dai, Hongli Yu, Qiying Yu, Xuefeng Li, Jiaze Chen, Hao Zhou, and Mingxuan Wang

Reference 2023

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T14:35:53.294734Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c47a4f8d-25df-4038-a11b-b53762be1fb8 · outbound

This paper cites an unresolved cited work.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Unresolved cited work

Reference 2024

Resolution: unresolved; raw_fallback, observed 2026-08-05T14:35:53.457163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a5b5ccbf-b8d3-43df-b2f1-c22b04f2d239 · outbound

This paper cites Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles.

Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-05T14:35:50.493578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation efa8a444-62f3-40aa-86a1-443620487913 · inbound

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs cites this paper.

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T10:13:15.414661Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6bf40004-0e52-4f1a-958a-38305042e275 · inbound

Learning to Build the Environment: Self-Evolving Reasoning RL via Verifiable Environment Synthesis cites this paper.

Learning to Build the Environment: Self-Evolving Reasoning RL via Verifiable Environment Synthesis Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions

Reference 41

Resolution: verified exact; arxiv_id, observed 2026-06-30T21:05:04.652492Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.