DUET improves RLVR by allocating tokens across both prompt selection and rollout length, outperforming full-budget baselines even when using only half the tokens.
Adaptive rollout allocation for online reinforcement learning with verifiable rewards
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
years
2026 6verdicts
UNVERDICTED 6roles
baseline 1polarities
baseline 1representative citing papers
HORA adaptively allocates rollouts using hit utility to improve Pass@K over compute-matched GRPO on math reasoning benchmarks while preserving Pass@1.
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
Dynamic Gradient Gating monitors lm_head gradient norms to safely reuse rollout batches in RLVR, achieving up to 2.93x sample efficiency and 2.14x wall-clock speedup across math, ALFWorld, WebShop, and QA tasks.
SORT turns all-wrong prompts into selective learning signals by weighting tokens more predictable under plan guidance from reference solutions, improving over GRPO on reasoning benchmarks especially for weaker models.
SimpleSearch-VL improves Qwen3-VL multimodal agent baselines by 15.8-16 points on average using 7K total training examples and reaches parity with Gemini-3-Pro on the 30B variant.
citing papers explorer
-
DUET: Optimize Token-Budget Allocation for Reinforcement Learning with Verifiable Rewards
DUET improves RLVR by allocating tokens across both prompt selection and rollout length, outperforming full-budget baselines even when using only half the tokens.
-
Where to Spend Rollouts: Hit-Utility Optimal Rollout Allocation for Group-Based RLVR
HORA adaptively allocates rollouts using hit utility to improve Pass@K over compute-matched GRPO on math reasoning benchmarks while preserving Pass@1.
-
Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
-
When to Stop Reusing: Dynamic Gradient Gating for Sample-Efficient RLVR
Dynamic Gradient Gating monitors lm_head gradient norms to safely reuse rollout batches in RLVR, achieving up to 2.93x sample efficiency and 2.14x wall-clock speedup across math, ALFWorld, WebShop, and QA tasks.
-
Selective Off-Policy Reference Tuning with Plan Guidance
SORT turns all-wrong prompts into selective learning signals by weighting tokens more predictable under plan guidance from reference solutions, improving over GRPO on reasoning benchmarks especially for weaker models.
-
SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search
SimpleSearch-VL improves Qwen3-VL multimodal agent baselines by 15.8-16 points on average using 7K total training examples and reaches parity with Gemini-3-Pro on the 30B variant.