AstraFlow decouples RL components into autonomous dataflow services to natively support multi-policy agentic LLM training, elastic scaling, and cross-region execution with 2.7x speedup on math, code, search, and AgentBench workloads.
hub Canonical reference
VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Tasks
Canonical reference. 73% of citing Pith papers cite this work as background.
50
Pith papers citing it
Background
73% of classified citations
abstract
We present VAPO, Value-based Augmented Proximal Policy Optimization framework for reasoning models., a novel framework tailored for reasoning models within the value-based paradigm. Benchmarked the AIME 2024 dataset, VAPO, built on the Qwen 32B pre-trained model, attains a state-of-the-art score of $\mathbf{60.4}$. In direct comparison under identical experimental settings, VAPO outperforms the previously reported results of DeepSeek-R1-Zero-Qwen-32B and DAPO by more than 10 points. The training process of VAPO stands out for its stability and efficiency. It reaches state-of-the-art performance within a mere 5,000 steps. Moreover, across multiple independent runs, no training crashes occur, underscoring its reliability. This research delves into long chain-of-thought (long-CoT) reasoning using a value-based reinforcement learning framework. We pinpoint three key challenges that plague value-based methods: value model bias, the presence of heterogeneous sequence lengths, and the sparsity of reward signals. Through systematic design, VAPO offers an integrated solution that effectively alleviates these challenges, enabling enhanced performance in long-CoT reasoning tasks.
hub tools
citation-role summary
background 11
method 3
dataset 1
citation-polarity summary
representative citing papers
AIS adaptively corrects non-stationary policy gradient bias in quantized LLM RL, matching BF16 performance while retaining 1.5-2.76x FP8 rollout speedup.
The cancellation hypothesis shows how rollout-level rewards produce token-level credit assignment in critic-free RL through cancellation of opposing signals on shared tokens, with empirical support and batching interventions that enhance performance.
POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
Reference-sampled weighted SFT with prompt-normalized Boltzmann weights induces the same policy as fixed-reference KL-regularized RLVR, with BOLT as the estimator and a finite one-shot error decomposition separating coverage, variance, and other terms.
GenAC introduces generative critics with chain-of-thought reasoning and in-context conditioning to improve value approximation and downstream RL performance in LLMs compared to value-based and value-free baselines.
SMTPO uses multi-task SFT to improve simulator feedback quality and RL with fine-grained rewards to optimize multi-turn preference reasoning in LLM-based conversational recommendation.
Positive-negative prompt pairing with weighted GRPO improves RLVR sample efficiency, raising AIME 2025 Pass@8 from 16.8 to 22.2 on Qwen2.5-Math-7B while matching large-scale training.
GPS trains a small model on optimization history to predict prompt difficulty and select intermediate-difficulty diverse batches, yielding better training efficiency, final performance, and test-time allocation than baselines on reasoning benchmarks.
A single LLM improves its own reasoning by self-distilling from privileged verified traces as teacher to its question-only student policy, outperforming off-policy distillation and RL on math benchmarks with better token efficiency.
MURPHY improves code generation pass rates by up to 6% through retrospective credit assignment on multi-turn feedback trees using max or mean reward propagation.
A 400k+ GPU-hour study shows RL scaling in LLMs follows predictable sigmoidal trajectories, with most design choices affecting efficiency rather than the performance asymptote, enabling accurate large-scale predictions via the ScaleRL recipe.
One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks.
SSOPD converts intra-group correct-wrong contrast into process supervision by distilling a teacher distribution from the shortest correct completion into prefixes of the longest wrong completion, improving GRPO on AIME and HMMT benchmarks.
Entropy polarity is a signed token-level quantity derived from a first-order approximation of entropy change that predicts whether RL updates expand or contract policy entropy in LLM fine-tuning, revealing an asymmetry between high- and low-probability tokens.
Seirênes trains LLMs via adversarial self-play to generate and overcome evolving distractions, producing gains of 7-10 points on math reasoning benchmarks and exposing blind spots in larger models.
Super-Linear Advantage Shaping (SLAS) introduces a non-linear geometric policy update for RL post-training of text-to-image models that reshapes the local policy space via advantage-dependent Fisher-Rao weighting to reduce reward hacking and improve performance over GRPO baselines.
OPT-BENCH trains LLMs on NP-hard optimization via quality-aware RLVR, achieving 93.1% success rate and 46.6% quality ratio on Qwen2.5-7B while outperforming GPT-4o and transferring gains to other domains.
AIPO adds active multi-agent consultation (Verify, Knowledge, Reasoning agents) plus custom importance sampling to RLVR training so LLMs expand their reasoning boundary and then operate without the agents.
HTPO introduces hierarchical token-level objective control in RLVR to balance exploration and exploitation by grouping tokens according to difficulty, correctness, and entropy, yielding up to 8.6% gains on AIME benchmarks over DAPO.
Mutual Reinforcement Learning allows heterogeneous LLMs to exchange experience through mechanisms like Peer Rollout Pooling, Cross-Policy GRPO Advantage Sharing, and Success-Gated Transfer, with outcome-level sharing identified as favorable on the stability-support trade-off.
GXPO approximates longer local lookahead in GRPO training via gradient extrapolation from two optimizer steps using three backward passes total, improving pass@1 accuracy by 1.65-5.00 points over GRPO and delivering up to 4x step speedup.
SAPO introduces segment-level policy optimization using a step-wise MDP abstraction to better align RL updates with reasoning structure in multi-modal LLM tasks.
LenVM models token-level remaining generation length as a bounded discounted value function derived from constant negative per-token rewards, providing a scalable proxy for generation horizon.
citing papers explorer
-
AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs
AstraFlow decouples RL components into autonomous dataflow services to natively support multi-policy agentic LLM training, elastic scaling, and cross-region execution with 2.7x speedup on math, code, search, and AgentBench workloads.
-
AIS: Adaptive Importance Sampling for Quantized RL
AIS adaptively corrects non-stationary policy gradient bias in quantized LLM RL, matching BF16 performance while retaining 1.5-2.76x FP8 rollout speedup.
-
The Cancellation Hypothesis in Critic-Free RL: From Outcome Rewards to Token Credits
The cancellation hypothesis shows how rollout-level rewards produce token-level credit assignment in critic-free RL through cancellation of opposing signals on shared tokens, with empirical support and batching interventions that enhance performance.
-
Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative Gradients
POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
-
Reference-Sampled Boltzmann Projection for KL-Regularized RLVR: Target-Matched Weighted SFT, Finite One-Shot Gaps, and Policy Mirror Descent
Reference-sampled weighted SFT with prompt-normalized Boltzmann weights induces the same policy as fixed-reference KL-regularized RLVR, with BOLT as the estimator and a finite one-shot error decomposition separating coverage, variance, and other terms.
-
Bringing Value Models Back: Generative Critics for Value Modeling in LLM Reinforcement Learning
GenAC introduces generative critics with chain-of-thought reasoning and in-context conditioning to improve value approximation and downstream RL performance in LLMs compared to value-based and value-free baselines.
-
User Simulator-Guided Multi-Turn Preference Optimization for Reasoning LLM-based Conversational Recommendation
SMTPO uses multi-task SFT to improve simulator feedback quality and RL with fine-grained rewards to optimize multi-turn preference reasoning in LLM-based conversational recommendation.
-
Beyond Variance: Prompt-Efficient RLVR via Rare-Event Amplification and Bidirectional Pairing
Positive-negative prompt pairing with weighted GRPO improves RLVR sample efficiency, raising AIME 2025 Pass@8 from 16.8 to 22.2 on Qwen2.5-Math-7B while matching large-scale training.
-
Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models
GPS trains a small model on optimization history to predict prompt difficulty and select intermediate-difficulty diverse batches, yielding better training efficiency, final performance, and test-time allocation than baselines on reasoning benchmarks.
-
Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models
A single LLM improves its own reasoning by self-distilling from privileged verified traces as teacher to its question-only student policy, outperforming off-policy distillation and RL on math benchmarks with better token efficiency.
-
MURPHY: Feedback-Aware GRPO with Retrospective Credit Assignment for Multi-Turn Code Generation
MURPHY improves code generation pass rates by up to 6% through retrospective credit assignment on multi-turn feedback trees using max or mean reward propagation.
-
The Art of Scaling Reinforcement Learning Compute for LLMs
A 400k+ GPU-hour study shows RL scaling in LLMs follows predictable sigmoidal trajectories, with most design choices affecting efficiency rather than the performance asymptote, enabling accurate large-scale predictions via the ScaleRL recipe.
-
Reinforcement Learning for Reasoning in Large Language Models with One Training Example
One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks.
-
Self-Supervised On-Policy Distillation for Reasoning Language Models
SSOPD converts intra-group correct-wrong contrast into process supervision by distilling a teacher distribution from the shortest correct completion into prefixes of the longest wrong completion, improving GRPO on AIME and HMMT benchmarks.
-
Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control
Entropy polarity is a signed token-level quantity derived from a first-order approximation of entropy change that predicts whether RL updates expand or contract policy entropy in LLM fine-tuning, revealing an asymmetry between high- and low-probability tokens.
-
Seir\^enes: Adversarial Self-Play with Evolving Distractions for LLM Reasoning
Seirênes trains LLMs via adversarial self-play to generate and overcome evolving distractions, producing gains of 7-10 points on math reasoning benchmarks and exposing blind spots in larger models.
-
Power Reinforcement Post-Training of Text-to-Image Models with Super-Linear Advantage Shaping
Super-Linear Advantage Shaping (SLAS) introduces a non-linear geometric policy update for RL post-training of text-to-image models that reshapes the local policy space via advantage-dependent Fisher-Rao weighting to reduce reward hacking and improve performance over GRPO baselines.
-
Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs
OPT-BENCH trains LLMs on NP-hard optimization via quality-aware RLVR, achieving 93.1% success rate and 46.6% quality ratio on Qwen2.5-7B while outperforming GPT-4o and transferring gains to other domains.
-
AIPO: Learning to Reason from Active Interaction
AIPO adds active multi-agent consultation (Verify, Knowledge, Reasoning agents) plus custom importance sampling to RLVR training so LLMs expand their reasoning boundary and then operate without the agents.
-
HTPO: Towards Exploration-Exploitation Balanced Policy Optimization via Hierarchical Token-level Objective Control
HTPO introduces hierarchical token-level objective control in RLVR to balance exploration and exploitation by grouping tokens according to difficulty, correctness, and entropy, yielding up to 8.6% gains on AIME benchmarks over DAPO.
-
Experience Sharing in Mutual Reinforcement Learning for Heterogeneous Language Models
Mutual Reinforcement Learning allows heterogeneous LLMs to exchange experience through mechanisms like Peer Rollout Pooling, Cross-Policy GRPO Advantage Sharing, and Success-Gated Transfer, with outcome-level sharing identified as favorable on the stability-support trade-off.
-
Gradient Extrapolation-Based Policy Optimization
GXPO approximates longer local lookahead in GRPO training via gradient extrapolation from two optimizer steps using three backward passes total, improving pass@1 accuracy by 1.65-5.00 points over GRPO and delivering up to 4x step speedup.
-
Segment-Aligned Policy Optimization for Multi-Modal Reasoning
SAPO introduces segment-level policy optimization using a step-wise MDP abstraction to better align RL updates with reasoning structure in multi-modal LLM tasks.
-
Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling
LenVM models token-level remaining generation length as a bounded discounted value function derived from constant negative per-token rewards, providing a scalable proxy for generation horizon.
-
V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
-
GRPO-VPS: Enhancing Group Relative Policy Optimization with Verifiable Process Supervision for Effective Reasoning
GRPO-VPS improves GRPO by using segment-wise conditional probabilities of the correct answer to supply process-level feedback, yielding up to 2.6-point accuracy gains and 13.7% shorter reasoning on math tasks.
-
HEALing Entropy Collapse: Enhancing Exploration in Few-Shot RLVR via Hybrid-Domain Entropy Dynamics Alignment
HEAL mitigates entropy collapse in few-shot RLVR by selectively adding general-domain data and aligning trajectory-level entropy dynamics, matching full-shot performance with 32 target samples.
-
Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillation
Lightning OPD is an offline on-policy distillation method that matches standard OPD performance at 4x efficiency by enforcing teacher consistency between SFT and distillation phases.
-
rePIRL: Learn PRM with Inverse RL for LLM Reasoning
rePIRL learns effective process reward models for LLM reasoning via a dual policy-PRM update process inspired by inverse RL, unifying online and offline methods with reported gains over prior approaches on math and coding datasets.
-
On the Non-decoupling of Supervised Fine-tuning and Reinforcement Learning in Post-training
SFT and RL cannot be decoupled in LLM post-training because each step increases the loss or lowers the reward of the prior step under KL and PL analyses.
-
From Ranking to Reasoning: Explainable Web API Recommendation via Semantic Reasoning
WAR-R1 combines special start/stop tokens in an LLM with supervised fine-tuning and GRPO reinforcement learning to deliver adaptive, explainable Web API recommendations that improve accuracy by up to 10.89% on ProgrammableWeb data.
-
When Importance Sampling Misallocates Credit: Asymmetric Ratios for Outcome-Supervised RL
The paper identifies that importance sampling ratios in outcome-supervised RL misallocate credit by creating unbalanced token updates, and introduces ASPO to correct the asymmetry for positive-advantage tokens.
-
Multiplayer Nash Preference Optimization
MNPO extends NLHF to multiplayer Nash games, inheriting equilibrium guarantees while showing empirical gains on instruction-following benchmarks under diverse preferences.
-
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Survey that defines agentic RL for LLMs via POMDPs, introduces a taxonomy of planning/tool-use/memory/reasoning capabilities and domains, and compiles open environments from over 500 papers.
-
InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling
InternBootcamp supplies 1000+ verifiable, auto-generated task environments across domains that enable task scaling to improve LLM reasoning, producing a 32B model with state-of-the-art results on the new Bootcamp-EVAL benchmark.
-
Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
Archer introduces response-level entropy normalization and differentiated clipping/KL regularization in RLVR to encourage exploration on reasoning tokens while stabilizing knowledge tokens, yielding gains in pass@1 and pass@K on reasoning benchmarks.
-
Writing-RL: Advancing Long-form Writing via Adaptive Curriculum Reinforcement Learning
Writing-RL applies adaptive curriculum RL with pairwise rewards and dynamic scheduling to enhance long-form writing in 7B LLMs over SFT baselines and shows generalization to long-input reasoning tasks.
-
AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning
AReaL decouples generation and training in LLM reinforcement learning to achieve up to 2.77x speedup with matched or better performance on math and code benchmarks.
-
ToolRL: Reward is All Tool Learning Needs
A principled reward design for tool selection and application in RL-trained LLMs delivers 17% gains over base models and 15% over SFT across benchmarks.
-
Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals
Proposes Near-boundary Stochastic Rescue (NSR) as a stochastic modification to clipping in RLVR that recovers near-boundary signals and yields gains over baselines like DAPO and GSPO.
-
PlexRL: Cluster-Level Orchestration of Serviceized LLM Execution for RLVR
PlexRL multiplexes unified LLM services across RLVR jobs at the cluster level to exploit anti-correlated idle times and reduce GPU-hour costs by up to 37.58% with minimal per-job overhead.
-
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
UI-TARS-2 reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld while attaining 59.8 mean normalized score on a 15-game suite through multi-turn RL and scalable data generation.
-
Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning
PODS applies max-variance down-sampling to GRPO rollouts in LLM RLVR, delivering at least 1.7x faster training to peak test accuracy on reasoning benchmarks.
-
Adaptive Negative Reinforcement for LLM Reasoning:Dynamically Balancing Correction and Diversity in RLVR
Adaptive scheduling of penalties over training time plus confidence-based weighting of mistakes improves LLM performance on math reasoning benchmarks compared to fixed-penalty negative reinforcement.
-
Sharpness-Guided Group Relative Policy Optimization via Probability Shaping
GRPO-SG is a sharpness-guided token-weighted variant of GRPO that downweights high-gradient tokens to stabilize optimization and improve generalization in reinforcement learning with verifiable rewards.
-
Skywork Open Reasoner 1 Technical Report
Skywork-OR1 uses RL on distilled CoT models to lift math and coding benchmark accuracy by 13-15 points while open-sourcing everything.
-
Seed1.5-VL Technical Report
Seed1.5-VL is a compact multimodal model that sets new records on dozens of vision-language benchmarks and outperforms prior systems on agent-style tasks.
-
Rethinking Agentic Reinforcement Learning In Large Language Models
The paper reviews conceptual foundations, methodological innovations, effective designs, critical challenges, and future directions for LLM-based Agentic Reinforcement Learning.
-
Reinforcement Learning from Human Feedback
The book introduces the origins, mathematical setup, and optimization stages of RLHF including reward modeling, reinforcement learning, rejection sampling, and direct alignment algorithms.
- STAPO: Stabilizing Reinforcement Learning for LLMs by Silencing Rare Spurious Tokens