Where Should RL Post-Training Compute Go? Model Size, Search, Learning, and Feedback

• URL: http://arxiv.org/abs/2607.13389v1
• Date: Wed, 15 Jul 2026 02:34:35 GMT
• Title: Where Should RL Post-Training Compute Go? Model Size, Search, Learning, and Feedback
• Abstract summary: Post-training resources are often summarized by a single total FLOP budget.<n>We introduce a FLOP-accounting framework for GRPO post-training that decomposes compute into rollout/search, policy-update/learning, and reward- or feedback-model evaluation.<n>Our results suggest that RL post-training papers should report total FLOPs together with how compute is divided among model size, search, learning, and feedback.
• Score: 1.5268922363885409
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement Learning (RL) post-training is increasingly used to adapt foundation models for reasoning, planning, and feedback-driven robot-learning pipelines, but constrained post-training resources are often summarized by a single total FLOP budget. We study the fixed-budget decision problem behind this practice: under the same post-training budget, should one use a larger policy, train a smaller policy longer, generate more rollout search, or spend compute on stronger reward feedback? We introduce a FLOP-accounting framework for GRPO post-training that decomposes compute into rollout/search, policy-update/learning, and reward- or feedback-model evaluation. Across LoRA-adapted Qwen2.5 policies, we find conditional allocation frontiers: the best observed allocation changes with model size, compute budget, reward system, and evaluation target. Same-FLOP model-size comparisons show that model choice and training allocation are coupled because larger policies consume more per-token compute and therefore buy fewer updates or rollouts under the same budget. Reward systems also change the accounting: rule-based rewards spend nearly all non-update compute on policy rollouts, while PRM-style feedback allocates a visible part of the budget to reward-model inference. We present RACE as a diagnostic pilot-grid protocol, not a guarantee of held-out improvement, for identifying allocation regimes before expensive validation runs; our results suggest that RL post-training papers should report total FLOPs together with how compute is divided among model size, search, learning, and feedback.

Related papers

• TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning [52.67756371231985]
  Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models.<n>We introduce Tree Rollout Allocation for Contrastive Exploration (TRACE), a unified rollout allocation framework.<n>Technically, TRACE allocates rollout budget to both prompt roots and intermediate prefixes that are most likely to yield mixed terminal rewards.
  arXiv  Detail & Related papers  (2026-06-09T17:16:03Z)
• Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training [34.81218483901967]
  Reinforcement learning is the dominant paradigm for post-training large language models.<n>Group-based policy optimization methods compute advantages from multiple rollouts per prompt.<n>Pilot-Commit is a budget-aware rollout allocation framework for group-based RL post-training.
  arXiv  Detail & Related papers  (2026-05-26T06:41:13Z)
• Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States [14.658333957936769]
  We introduce Policy Optimization with Internal State Value Estimation.<n>It obtains a baseline at negligible cost by using the policy model's internal signals already computed during the policy forward pass.<n>A lightweight probe predicts the expected verifiable reward from the hidden states of the prompt and generated trajectory.
  arXiv  Detail & Related papers  (2026-05-08T10:49:36Z)
• Where to Spend Rollouts: Hit-Utility Optimal Rollout Allocation for Group-Based RLVR [25.35547462790362]
  Group-based policy optimization methods typically allocate a fixed number of rollouts to every prompt.<n>We introduce hit utility, the posterior probability that at least one rollout in a proposed additional allocation for a prompt will be correct.<n>We propose Hit-Utility Optimal Rollout Allocation (HORA), a learning-free rollout allocation policy.
  arXiv  Detail & Related papers  (2026-05-08T01:42:25Z)
• Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning [79.88942231770629]
  Reinforcement learning (RL) has become a central post-training tool for improving the reasoning abilities of large language models (LLMs)<n>We formalize rollout pipelines with unified notation and introduce Generate-Filter-Control-Replay (GFCR)<n>We synthesize methods spanning RL with verifiable rewards, process supervision, judge-based gating, guided and tree/segment rollouts, adaptive compute allocation, early-exit and partial rollouts, throughput optimization, and replay/recomposition for self-improvement.
  arXiv  Detail & Related papers  (2026-04-08T00:53:29Z)
• Jackpot: Optimal Budgeted Rejection Sampling for Extreme Actor-Policy Mismatch Reinforcement Learning [31.843129392507716]
  Jackpot is a framework that integrates a unified training objective that jointly updates the policy and rollout models.<n>Our theoretical analysis shows that OBRS consistently moves the rollout distribution closer to the target distribution under a controllable acceptance budget.
  arXiv  Detail & Related papers  (2026-02-05T18:57:01Z)
• $V_0$: A Generalist Value Model for Any Policy at State Zero [80.7505802128501]
  Policy methods rely on a baseline to measure the relative advantage of an action.<n>This baseline is typically estimated by a Value Model (Critic) often as large as the policy model itself.<n>We propose a Generalist Value Model capable of estimating the expected performance of any model on unseen prompts.
  arXiv  Detail & Related papers  (2026-02-03T14:35:23Z)
• FlowRL: Matching Reward Distributions for LLM Reasoning [69.88820066093798]
  We propose FlowRL: matching the full reward distribution via flow balancing instead of maximizing rewards in large language model (LLM) reinforcement learning (RL)<n>We transform scalar rewards into a normalized target distribution using a learnable partition function, and then minimize the reverse KL divergence between the policy and the target distribution.
  arXiv  Detail & Related papers  (2025-09-18T17:56:36Z)
• Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning [55.15106182268834]
  Reinforcement learning with verifiable rewards (RLVR) has emerged as the leading approach for enhancing reasoning capabilities in large language models.<n>It faces a fundamental compute and memory asymmetry: rollout generation is embarrassingly parallel and memory-light, whereas policy updates are communication-heavy and memory-intensive.<n>We introduce PODS (Policy Optimization with Down-Sampling), which decouples rollout generation from policy updates by training only on a strategically selected subset of rollouts.
  arXiv  Detail & Related papers  (2025-04-18T17:49:55Z)
• Lean and Mean: Decoupled Value Policy Optimization with Global Value Guidance [52.65461207786633]
  Policy-based Reinforcement Learning from Human Feedback is essential for aligning large language models with human preferences.<n>It requires joint training of an actor and critic with a pretrained, fixed reward model for guidance.<n>We propose textbfDecoupled Value Policy Optimization (DVPO), a lean framework that replaces traditional reward modeling with a pretrained emphglobal value model (GVM)
  arXiv  Detail & Related papers  (2025-02-24T08:11:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.