PAIR: Pairwise-Aware Inclusion Reweighting for Adaptive Rollout Allocation in RLVR

• URL: http://arxiv.org/abs/2608.11368v2
• Date: Thu, 13 Aug 2026 13:43:42 GMT
• Title: PAIR: Pairwise-Aware Inclusion Reweighting for Adaptive Rollout Allocation in RLVR
• Abstract summary: Reinforcement learning with verifiable rewards spends most of its compute generating groups of long reasoning trajectories.<n>Recent allocators reduce this cost by assigning budgets to prompts, rollouts, or tokens according to a pointwise notion of difficulty or utility.<n>We identify a statistical mismatch: the unclipped leave-one-out group-relative score gradient is not a sum of independent point contributions, but a second-order U-statistic over pairs of rollouts.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with verifiable rewards (RLVR) spends most of its compute generating groups of long reasoning trajectories. Recent allocators reduce this cost by assigning budgets to prompts, rollouts, or tokens according to a pointwise notion of difficulty or utility. We identify a statistical mismatch: the unclipped leave-one-out group-relative score gradient is not a sum of independent point contributions, but a second-order U-statistic over pairs of rollouts. Completing one rollout therefore reveals contrast with every other completed rollout, and adaptive endpoint selection changes which pair terms are observable. We introduce PAIR (Pairwise-Aware Inclusion Reweighting), which treats short rollout prefixes as vertices and pair-gradient terms as edges of a contrast graph. A prefix-only predictor estimates correctness and remaining token cost; a convex design chooses positive continuation probabilities under an expected suffix-token budget; and each edge induced by completed vertices is inverse-weighted by its logged joint inclusion probability. Under conditionally independent on-policy rollouts and an unclipped, unstandardized objective, the resulting estimator is design-unbiased for the complete candidate-pair gradient. Across compute-matched RLVR runs on Qwen3-1.7B/4B, PAIR improves average accuracy by +1.2 and +1.4 over the strongest pointwise allocator while using 51% and 52% fewer generated tokens than full-group GRPO. A frozen-population estimator audit confirms that unweighted adaptive selection is biased, whereas pair-inclusion correction recovers the complete-pair target at matched suffix cost.

Related papers

• Convex-Hull-Neighborhood Smooth Dual Generalization: Controlling Local Correction Propagation in Offline RL [7.803284875836737]
  bootstrapping can amplify estimation errors in offline reinforcement learning.<n>We propose Convex Hull Neighborhood Smooth Dual Generalization, which expresses the Bellman backup as an in-sample value target plus a CHN-local correction.<n> Experiments on Gym-MuJoCo and AntMaze show strong aggregate performance and stable value estimation.
  arXiv  Detail & Related papers  (2026-08-04T04:30:02Z)
• 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)
• Instance-Optimal Estimation with Multiple LLM Judges on a Budget [84.31744861038106]
  We formalize this question as *budgeted heteroskedastic multi-judge estimation*.<n>Given $K$ prompt-response pairs, $J$ judges with known costs, and unknown query-judge variances, the goal is to estimate a bounded score vector while minimizing an $ell_p$-error.<n>We prove that EST-IVWE matches the oracle IVWE rate up to lower-order terms in the budget.
  arXiv  Detail & Related papers  (2026-05-22T08:26:08Z)
• When Individually Calibrated Models Become Collectively Miscalibrated [3.556355987197792]
  We show that individually calibrated predictors can become collectively miscalibrated when their predictions interact strategically.<n>We prove that under Brier-score-based aggregation with positively correlated beliefs, each agent's individually optimal report systematically underestimates the positive-class probability.<n>In contrast, VCG-based aggregation aligns incentives by rewarding marginal contribution, achieving dominant-strategy incentive compatibility and near-optimal performance.
  arXiv  Detail & Related papers  (2026-05-14T05:25:16Z)
• Revisiting Reinforcement Learning with Verifiable Rewards from a Contrastive Perspective [10.958642517467721]
  We present a framework for Contrastive Sequence-level Policy Optimization in RLVR.<n>ConSPO replaces GRPO's clipped ratio-based scores with length-normalized sequence log-probabilities.<n>ConSPO consistently outperforms several strong RLVR baselines on challenging mathematical reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-05-13T04:02:36Z)
• AdaPaD: Adaptive Parallel Deflation for PEFT with Self-Correcting Rank Discovery [10.176501817419371]
  We present AdaPaD (Adaptive Parallel Deflation), which trains all rank-1 components simultaneously.<n>We prove that every component's error decays exponentially after a warm-up period.<n>AdaPaD is competitive with adaptive-rank LoRA baselines on GLUE with DeBERTaV3-base at matched parameter budgets.
  arXiv  Detail & Related papers  (2026-05-11T15:44:13Z)
• Balanced Aggregation: Understanding and Fixing Aggregation Bias in GRPO [70.38763678943648]
  Reinforcement learning with verifiable rewards (RLVR) has become a central paradigm for improving reasoning and code generation in large language models.<n>Standard GRPO uses sequence aggregation, while recent work has advocated token aggregation as a better alternative.<n>We show that these two rules induce different optimization biases; token aggregation introduces sign-length coupling, while sequence aggregation implicitly downweights longer responses.
  arXiv  Detail & Related papers  (2026-04-14T09:48:46Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• ZIP-RC: Optimizing Test-Time Compute via Zero-Overhead Joint Reward-Cost Prediction [57.799425838564]
  We present ZIP-RC, an adaptive inference method that equips models with zero-overhead inference-time predictions of reward and cost.<n> ZIP-RC improves accuracy by up to 12% over majority voting at equal or lower average cost.
  arXiv  Detail & Related papers  (2025-12-01T09:44:31Z)
• Semiparametric conformal prediction [79.6147286161434]
  We construct a conformal prediction set accounting for the joint correlation structure of the vector-valued non-conformity scores.<n>We flexibly estimate the joint cumulative distribution function (CDF) of the scores.<n>Our method yields desired coverage and competitive efficiency on a range of real-world regression problems.
  arXiv  Detail & Related papers  (2024-11-04T14:29:02Z)

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.