Paper Citation Record · LEDGER

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training

As of 6 August 2026, this Paper Citation Record lists 23 of 23 outbound references and 0 inbound Pith citation observations for arXiv:2605.26606.

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

pith.paper-citation-record.v1
2605.26606 v1

Coverage vector

measured 23 of 23 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-29T19:47:09.817243Z

measured 23 of 23 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-06T06:34:29.942622+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

23 of 23 outbound references displayed

verified exact17
verified fuzzy0
unresolved2
parse uncertain0
malformed identifier1
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation c7a46cd7-7851-4120-bc6c-79ba15bf77a2 · outbound

This paper cites arXiv preprint arXiv:2503.18929 , year=.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training arXiv preprint arXiv:2503.18929 , year=

Reference 1

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T19:53:55.801831Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation bc1ff799-a1ca-44df-85c5-e36db16bccdd · outbound

This paper cites Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901,.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901,

Reference 2

Resolution: unresolved; no resolver link, observed 2026-06-29T19:47:09.817243Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8bf88d01-7253-4077-8366-29b2d1bb9843 · outbound

This paper cites arXiv preprint arXiv:2510.01135 , year=.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training arXiv preprint arXiv:2510.01135 , year=

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.804948Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ea838eab-8e97-476f-8d80-a2736669e7f2 · outbound

This paper cites DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.807818Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation b8f67a05-ae1b-49c7-9cb0-56028ac47fa2 · outbound

This paper cites Vcrl: Variance-based curriculum reinforcement learning for large language models.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Vcrl: Variance-based curriculum reinforcement learning for large language models

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.798410Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation d31b275c-48fa-449e-9618-c02a80b45c53 · outbound

This paper cites Tinker, 2025.https://thinkingmachines.ai/tinker/.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Tinker, 2025.https://thinkingmachines.ai/tinker/

Reference 6

Resolution: unresolved; no resolver link, observed 2026-06-29T19:47:09.817243Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49ee130d-b97a-481a-bbc6-aad4c82d7f60 · outbound

This paper cites Knapsack rl: Unlocking exploration of llms via optimizing budget allocation.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Knapsack rl: Unlocking exploration of llms via optimizing budget allocation

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.792753Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2a7091f0-4665-48a8-985e-2e89b0a6701c · outbound

This paper cites Cppo: Accelerating the training of group relative policy optimization-based reasoning models.arXiv preprint arXiv:2503.22342.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Cppo: Accelerating the training of group relative policy optimization-based reasoning models.arXiv preprint arXiv:2503.22342

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.748045Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 4c01d6d6-43f7-46fb-98a3-f07d1c7e050e · outbound

This paper cites Understanding R1-Zero-Like Training: A Critical Perspective.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Understanding R1-Zero-Like Training: A Critical Perspective

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.750872Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation fca16e96-44c4-4d55-97a6-5abccc197354 · outbound

This paper cites Yun Qu, Qi Wang, Yixiu Mao, Vincent Tao Hu, Bj¨orn Ommer, and Xiangyang Ji.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Yun Qu, Qi Wang, Yixiu Mao, Vincent Tao Hu, Bj¨orn Ommer, and Xiangyang Ji

Reference 10

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T19:53:55.754059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 1627ce8b-ed7d-4c44-85cb-af20c9fc0f63 · outbound

This paper cites Proximal Policy Optimization Algorithms.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Proximal Policy Optimization Algorithms

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.773627Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 46bd31e0-233b-4137-a939-e71ad415c689 · outbound

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

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.804385Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation cbab1d9f-87b4-4cd9-8e91-b32ecddde4a7 · outbound

This paper cites HybridFlow: A Flexible and Efficient RLHF Framework.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training HybridFlow: A Flexible and Efficient RLHF Framework

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.810454Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation a4ec028c-7bf4-477f-9a6d-f5b3532c8d46 · outbound

This paper cites Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.807528Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 3ca7aa93-bd09-4601-9abd-9e0159acd6aa · outbound

This paper cites LlamaRL: A Distributed Asynchronous Reinforcement Learning Framework for Efficient Large-scale LLM Training.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training LlamaRL: A Distributed Asynchronous Reinforcement Learning Framework for Efficient Large-scale LLM Training

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.792604Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2daae097-2b59-42e0-a985-0b65c6a84215 · outbound

This paper cites Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.795093Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation f2b379f6-5468-45d0-a1de-67dbfc86a075 · outbound

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

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.795243Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 729a0b42-3aba-4068-b985-aed5c8184b19 · outbound

This paper cites Qwen3 Technical Report.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Qwen3 Technical Report

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.811407Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 36f0d166-0b90-4c03-9c82-f3d3b7ca186f · outbound

This paper cites DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.785979Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation dc0b3142-4e43-48af-a3ee-e1c664c391b6 · outbound

This paper cites DAPO: An Open-Source LLM Reinforcement Learning System at Scale.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.782296Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 37914b51-0026-4c78-83a7-9a0a1fb415f4 · outbound

This paper cites Speed-rl: Faster training of reasoning models via online curriculum learning.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Speed-rl: Faster training of reasoning models via online curriculum learning

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-06-29T19:53:55.801652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 942d9897-902c-48e5-9eb4-31e12c4ae2cd · outbound

This paper cites Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts

Reference 22

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T19:53:55.782288Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 0db1c8d8-bc8f-4af4-aaee-27cb384a58c2 · outbound

This paper cites Rollouts Trained.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training Rollouts Trained

Reference 23

Resolution: malformed identifier; no resolver link, observed 2026-06-29T19:47:09.817243Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.