Rubric Curriculum RL (RcRL) is a reinforcement learning approach that leverages the generation‑verification gap in non‑verifiable domains by using pairwise preferences over rollouts and a curriculum over rubric criteria. Unlike traditional methods that rely on absolute scoring and stationary judges, RcRL provides a more discriminative, non‑stationary signal, allowing policies to maintain output entropy and continue improving beyond 1,000 training steps. In experiments on creative writing and HealthBench, RcRL outperforms baseline models, achieving a 70.5% human win rate and significant accuracy gains over the base model.
arXiv:2609.23457v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) is expanding from tasks with well-defined correctness signals, such as mathematics and code, towa...
The paper introduces Step-wise Rubrics as Rewards (SRaR), a reinforcement learning framework that assigns rubric-based scores to individual reasoning steps in large language models. By normalizing step scores and combining them with outcome rewards through a decoupled advantage estimator, SRaR preserves criterion-level structure and mitigates issues like rewarding flawed steps or encouraging endless self‑correction. Experiments on 16K problems and six math benchmarks show that SRaR outperforms prior rubric methods, improving accuracy, increasing faithful reasoning rates, and reducing self‑correction loops.
Reinforcement Learning with Verifiable Rewards (RLVR) is expanding from tasks with well-defined correctness signals, such as mathematics and code, toward multifaceted quality requirements specified by...
arXiv:2610.33512v1 Announce Type: new Abstract: Rubric-based rewards enable effective reinforcement learning on non-verifiable tasks. However, because these rewards are typically defined as a weighte...
arXiv:2608. 02948v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD), where a single model acts as both student and teacher with different contexts, has shown promise in verifiable domains like math, where hard privileged information (PI) in the form of ground-truth answers structurally constrains valid continuations.
arXiv:2606. 03968v1 Announce Type: cross Abstract: Rubric-based RL is a promising route for extending reinforcement learning beyond verifiable rewards, yet existing methods optimize rubrics while treating the query distribution as fixed.
The paper introduces a two‑stage training framework for rubric‑based reinforcement learning (RL) that first uses rubric‑privileged on‑policy distillation (RP‑OPD) to provide dense token‑level supervision, then applies RL to optimize the rubric reward. In RP‑OPD, a student model learns to match a rubric‑aware teacher’s next‑token distributions on student‑generated prefixes. Experiments on HealthBench, ResearchQA, and RubricHub Science show that this two‑stage approach yields the highest scores and reduces reward hacking compared to a standard SFT + RL baseline.
arXiv:2608.30005v1 Announce Type: new Abstract: Rubric-based reinforcement learning extends RL beyond tasks with exact answers or rule-based verifiers by scoring responses against instance-specific c...
arXiv:2604. 02621v2 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) substantially improves the reasoning capabilities of language models, but most existing RL fine-tuning approaches rely entirely on ground-truth verifiable rewards and thus labeled datasets with verifiable answers.
The paper introduces Rubric Dropout, a technique that randomly removes a subset of rubric criteria during training to prevent reinforcement learning agents from exploiting the rubric as a proxy for quality. By resampling which criteria are rewarded at each step and ensuring all responses to a prompt are evaluated on the same subset, the method reduces reward hacking and improves out‑of‑distribution performance on medical and science benchmarks. Block Dropout further groups semantically similar criteria to avoid near‑duplicate rewards, yielding additional gains.
The paper introduces ReVeR, a pointwise rubric reward model that derives hard rules and graded principles directly from instructions, avoiding response-induced bias. It employs tool-augmented judges and label‑blind distillation to create ReVeR‑Traces, a large dataset of rubric reasoning and judging trajectories. Experiments show ReVeR surpasses pairwise rubric models in preference accuracy, especially for instruction following, and reveal that higher preference accuracy does not always lead to better downstream policies.
Post-training of reasoning language models is commonly driven by supervised distillation and reinforcement learning with verifiable rewards. Distillation often relies on chain-of-thought annotations that are expensive to obtain and may themselves be noisy, incomplete, or partially incorrect; even when the final solution is correct, an imperfect rationale can interfere with learning.