Claude 3 Opus strategically fakes alignment by complying with harmful requests only during simulated training to preserve its preference for refusing them afterward.
mega hub Mixed citations
Proximal Policy Optimization Algorithms
Mixed citation behavior. Most common role is background (52%).
1645
Pith papers citing it
Background
52% of classified citations
abstract
We propose a new family of policy gradient methods for reinforcement learning, which alternate between sampling data through interaction with the environment, and optimizing a "surrogate" objective function using stochastic gradient ascent. Whereas standard policy gradient methods perform one gradient update per data sample, we propose a novel objective function that enables multiple epochs of minibatch updates. The new methods, which we call proximal policy optimization (PPO), have some of the benefits of trust region policy optimization (TRPO), but they are much simpler to implement, more general, and have better sample complexity (empirically). Our experiments test PPO on a collection of benchmark tasks, including simulated robotic locomotion and Atari game playing, and we show that PPO outperforms other online policy gradient methods, and overall strikes a favorable balance between sample complexity, simplicity, and wall-time.
hub tools
citation-role summary
background 155
method 113
baseline 15
dataset 4
citation-polarity summary
claims ledger
- abstract We propose a new family of policy gradient methods for reinforcement learning, which alternate between sampling data through interaction with the environment, and optimizing a "surrogate" objective function using stochastic gradient ascent. Whereas standard policy gradient methods perform one gradient update per data sample, we propose a novel objective function that enables multiple epochs of minibatch updates. The new methods, which we call proximal policy optimization (PPO), have some of the benefits of trust region policy optimization (TRPO), but they are much simpler to implement, more ge
authors
mega hub controls
export citing contexts JSON
export graph JSON
export full bundle JSON
open full Pith review
annotated reader queued
Recognition alignment
counterfactual ablation
co-cited works
representative citing papers
GRPO's group-mean baseline assigns identical advantages to all tokens under output-only rewards, inducing gradient sparsity and an intrinsic rank-2 structure proven from the zero-sum constraint and confirmed by SVD on Nemotron-4B gradients.
Negative narrative immersion causes 12-31% drops in LLM moral accuracy and produces structured shifts that appear in downstream applications.
RL agent for online LHC trigger threshold tuning improves in-tolerance intervals by 28-56% on Monte Carlo and real CMS data without fine-tuning.
A reward-free representation learning pipeline for offline PbRL achieves better preference efficiency than standard two-stage baselines by connecting RFRL concepts to preference data.
Dynamic isotropy, quantifying uniform center-of-mass acceleration capability, improves robot performance and enables omnidirectional locomotion, terrain traversal, and failure resilience in a spherical robot design.
AtomComposer uses online RL with multi-composition training to discover up to 10x more valid 3D isomers on unseen chemical formulas than single-composition baselines.
Agent-BRACE improves LLM agent performance on long-horizon partially observable tasks by 5.3-14.5% through a decoupled belief state of verbalized atomic claims with certainty labels that keeps context length constant.
SimWorld Studio deploys an evolving coding agent to create adaptive 3D environments that co-evolve with embodied learners, delivering 18-point success-rate gains over fixed environments in navigation benchmarks.
ReLibra uses pre-known token-to-expert routing from RL rollouts to perform inter-batch expert reordering and intra-batch replication, delivering up to 1.6x higher throughput than Megatron-LM and 1.2x over oracle-equipped EPLB while staying within 6-10% of an ideal balanced baseline.
Weak-to-strong generalization is nearly inevitable in linear logistic regression for most student-teacher pairs without any model capacity mismatch.
Observation and action delays are formally equivalent in cooperative Dec-POMDPs, yielding identical optimal solutions and enabling zero-shot transfer, though learning dynamics differ due to credit assignment and operational constraints.
A language-game framework enables dialogue with dynamical systems such as GRNs by treating their frozen dynamics as an RL policy core, using an LM to route prompts so the system responds through its own behavior without parameter changes.
RefereeBench shows that even the strongest video MLLMs reach only around 60% accuracy on multi-sport refereeing tasks and struggle with rule application and temporal grounding.
OP-GRPO is the first off-policy GRPO method for flow-matching models that reuses trajectories via replay buffer and importance sampling corrections, matching on-policy performance with 34.2% of the training steps.
User-turn generation reveals that LLMs' interaction awareness is largely decoupled from task accuracy, remaining near zero in deterministic settings even as accuracy scales to 96.8% on GSM8K.
Derives an exact unbiased policy gradient for RL post-training of diffusion LLMs via entropy-guided step selection and one-step denoising rewards, achieving state-of-the-art results on coding and logical reasoning benchmarks.
A certified gradient-based method for contact-rich manipulation that quantifies smoothing-induced errors via set-valued discrepancies and incorporates them into analytical reachable sets for robust affine feedback policies.
First in-orbit demonstration of a DRL-trained AI satellite attitude controller that performs robust inertial pointing after sim-to-real transfer.
Develops and tests the first effective safeguard for analytic gradient-based provably safe RL, showing safe training on three control tasks without performance loss.
Flow-GRPO is the first online RL method for flow matching models, raising GenEval accuracy from 63% to 95% and text-rendering accuracy from 59% to 92% with little reward hacking.
SWE-Gym supplies 2438 executable real-world Python tasks to train SWE agents and verifiers, yielding up to 19% gains and new open-weight SOTA of 32% on SWE-Bench Verified.
BEHAVIOR-1K introduces a benchmark of 1,000 human everyday activities in realistic simulated scenes together with the OMNIGIBSON physics simulator to evaluate embodied AI.
ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.
citing papers explorer
-
STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering
STRIVE stabilizes RL for video QA by creating spatiotemporal video variants and using importance-aware sampling, yielding consistent gains over baselines on six benchmarks.
-
TestDecision: Sequential Test Suite Generation via Greedy Optimization and Reinforcement Learning
By proving test suite coverage is monotone submodular and training LLMs with RL to maximize marginal gains, TestDecision improves branch coverage 38-52% and bug detection up to 95% over base models on ULT and LiveCodeBench.
-
Reliability-Aware Geometric Fusion for Robust Audio-Visual Navigation
RAVN improves audio-visual navigation by learning audio-derived reliability cues via an Acoustic Geometry Reasoner and using them to modulate visual features through Reliability-Aware Geometric Modulation.
-
Heterogeneous Self-Play for Realistic Highway Traffic Simulation
PHASE uses heterogeneous self-play and context-conditioned policies to achieve realistic, zero-shot highway traffic simulation that outperforms traditional rule-based and self-play models on real-world datasets.
-
MemFactory: Unified Inference & Training Framework for Agent Memory
MemFactory is a new unified modular framework for memory-augmented LLM agent inference and training that integrates GRPO and reports up to 14.8% relative gains on MemAgent evaluations.
-
Learning Unified Control of Intrinsic Nonlinear Spin Dynamics in Atomic Qudits for Magnetometry
Reinforcement learning stabilizes more than 4 dB of fixed-axis spin squeezing under continuous nonlinear Zeeman evolution in the f=21/2 manifold of 161Dy, yielding a single-atom sensitivity of 13.9 pT/sqrt(Hz) that is 3 dB beyond the standard quantum limit.
-
Competitor-aware Race Management for Electric Endurance Racing
A bi-level game-theoretic optimal control plus reinforcement learning framework enables competitor-aware energy management and pit-stop scheduling that exploits aerodynamic drafting in simulated electric endurance races.
-
AdaRubric: Task-Adaptive Rubrics for Reliable LLM Agent Evaluation and Reward Learning
AdaRubric adaptively generates task-specific rubrics via LLM, scores agent trajectories with per-dimension confidence weighting, and produces filtered DPO pairs that raise human correlation to Pearson r=0.79 and downstream task success by 6.8-8.5%.
-
Delightful Distributed Policy Gradient
Delightful Policy Gradient gates updates with advantage times surprisal to suppress rare failures while preserving rare successes in distributed RL with stale or buggy data.
-
PersonaVLM: Long-Term Personalized Multimodal LLMs
PersonaVLM adds memory extraction, multi-turn retrieval-based reasoning, and personality inference to multimodal LLMs, yielding 22.4% gains on a new long-term personalization benchmark and outperforming GPT-4o.
-
Teaching an Agent to Sketch One Part at a Time
A multi-modal LM agent is trained to produce vector sketches part-by-part via supervised fine-tuning and process-reward RL on the new ControlSketch-Part dataset with automatic part annotations.
-
Optimizing Resource-Constrained Non-Pharmaceutical Interventions for Multi-Cluster Outbreak Control Using Hierarchical Reinforcement Learning
Hierarchical RL with a global cost controller and local marginal-value policies outperforms RMAB and heuristic baselines by 20-30% in simulated multi-cluster SARS-CoV-2 control.
-
VC-Soup: Value-Consistency Guided Multi-Value Alignment for Large Language Models
VC-Soup uses a cosine-similarity consistency metric to filter data, trains value-consistent policies, and applies linear merging with Pareto filtering to improve multi-value LLM alignment trade-offs.
-
Simulation Distillation: Pretraining World Models in Simulation for Rapid Real-World Adaptation
SimDist pretrains world models in simulation and adapts them to real-world robots by updating only the latent dynamics model, enabling rapid improvement on contact-rich tasks where prior methods fail.
-
Learning to Assist: Physics-Grounded Human-Human Control via Multi-Agent Reinforcement Learning
AssistMimic is the first multi-agent RL method that successfully tracks assistive human-human interaction motions in simulation by using partner-aware policies, single-agent initialization, dynamic reference retargeting, and contact-promoting rewards.
-
Hindsight-Anchored Policy Optimization: Turning Failure into Feedback in Sparse Reward Settings
HAPO adds a hindsight-anchored SSI operator with Thompson gating to GRPO-style RLVR, achieving asymptotic consistency that recovers unbiased on-policy gradients as the policy improves.
-
MICA: Multi-granularity Intertemporal Credit Assignment for Long-Horizon Emotional Support Dialogue
MICA combines incremental per-turn distance rewards and Monte Carlo returns from a shared potential function over user support states to create a mixed advantage signal that enables stable multi-turn RL optimization for emotional support dialogues.
-
LucidNFT: LR-Anchored Multi-Reward Preference Optimization for Flow-Based Real-World Super-Resolution
LucidNFT combines a new LR-referenced consistency reward, decoupled normalization, and a real-degradation dataset to improve perceptual quality in flow-matching super-resolution while preserving input fidelity.
-
Watch Your Step: Learning Semantically-Guided Locomotion in Cluttered Environment
SemLoco is a reinforcement learning system that integrates semantic understanding with foothold planning to let legged robots navigate cluttered environments without stepping on sensitive low-lying objects.
-
Learning Vision-Based Omnidirectional Navigation: A Teacher-Student Approach Using Monocular Depth Estimation
A distilled student policy using monocular depth estimation from cameras outperforms a 2D LiDAR teacher policy in navigating complex 3D obstacles while running fully onboard a Jetson Orin.
-
HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents
HiMAC decomposes LLM agent tasks into macro planning and micro execution using critic-free hierarchical RL and iterative co-evolution, outperforming baselines on ALFWorld, WebShop, and Sokoban.
-
GeoWorld: Geometric World Models
GeoWorld applies hyperbolic geometry to JEPA world models and introduces geometric reinforcement learning, reporting modest success-rate gains of ~3% and ~2% on 3- and 4-step planning tasks versus V-JEPA 2.
-
Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving
The paper introduces Hyper Diffusion Planner (HDP), a diffusion-based E2E AD framework that identifies insights on loss space, trajectory representation and data scaling, adds RL post-training, and reports 10x performance gains over 200 km of real-world testing across 6 scenarios.
-
Space Syntax-guided Post-training for Residential Floor Plan Generation
SSPT turns space-syntax integration metrics into post-training feedback signals that improve public-space dominance and functional hierarchy in AI-generated residential floor plans.
-
RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection
RL-RIG uses a generate-reflect-edit loop with reinforcement learning to improve spatial accuracy in image generation, reporting up to 11% gains over prior open-source models on scene-graph metrics.
-
TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs
TempoNet uses a slack-quantized Transformer with deep Q-learning and sparse attention to improve deadline fulfillment rates over traditional and neural schedulers in mixed-criticality real-time workloads.
-
One Hand to Rule Them All: Canonical Representations for Unified Dexterous Manipulation
A unified parameter space and canonical URDF enable cross-embodiment dexterous grasping policies with 81.9% zero-shot success on unseen hands like the 3-finger LEAP Hand.
-
RIDER: 3D RNA Inverse Design with Reinforcement Learning-Guided Diffusion
RIDER improves RNA 3D structural similarity by over 100% using RL-guided diffusion and discovers non-native sequence designs.
-
Multi-Modal Multi-Agent Reinforcement Learning for Radiology Report Generation
MARL-Rad trains region-specific and global agents with reinforcement learning on clinical rewards to produce more accurate radiology reports than prior methods on MIMIC-CXR and IU X-ray datasets.
-
HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Model
HAIC enables robust humanoid interactions with underactuated objects by predicting their dynamics from proprioceptive history and using a world model for adaptive control.
-
Learning to Configure Agentic AI Systems
ARC learns per-query agent configurations via a lightweight hierarchical SMDP policy, delivering 31.3% higher reasoning accuracy, 13.95% higher tool-use accuracy, and doubled success on an agent benchmark compared to budget-matched baselines.
-
RISE: Self-Improving Robot Policy with Compositional World Model
RISE combines a controllable dynamics model and progress value model into a closed-loop self-improving pipeline that updates robot policies entirely in imagination, reporting over 35% absolute gains on three real-world tasks.
-
CMAD: Cooperative Multi-Agent Diffusion via Stochastic Optimal Control
CMAD formulates compositional generation as cooperative stochastic optimal control among pre-trained diffusion models, validated on conditional MNIST against a gradient-guidance baseline.
-
Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning
LifeLong-RFT applies chunking-level on-policy reinforcement learning with Quantized Action Consistency Reward, Continuous Trajectory Alignment Reward, and Format Compliance Reward to fine-tune VLA models, achieving a 22% average success rate gain over supervised fine-tuning on the LIBERO benchmark's
-
The Role of Learning in Attacking ML-based Network Intrusion Detection
Reinforcement learning agents achieve up to 58.1% attack success on ML network intrusion detectors at 0.31 ms per attack, delivering over 1000X higher throughput than gradient-based methods while working directly on non-differentiable models.
-
TeleGate: Whole-Body Humanoid Teleoperation via Gated Expert Selection with Motion Prior
TeleGate achieves high-precision real-time whole-body teleoperation of humanoid robots by dynamically gating between expert policies and using a VAE motion prior to infer future intent from history, outperforming distillation baselines on dynamic motions with only 2.5 hours of mocap data.
-
Phase-Aware Policy Learning for Skateboard Riding of Quadruped Robots via Feature-wise Linear Modulation
PAPL uses phase-conditioned FiLM layers in RL networks to create a unified policy for quadruped robots to ride skateboards by capturing phase-dependent behaviors while sharing knowledge across phases.
-
TwinRL: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation
TwinRL expands RL exploration via digital twin reconstruction and twin RL warm-up to guide real-world learning, reaching near-100% success with 20 minutes of on-robot time across four tasks.
-
Robust Policy Optimization to Prevent Catastrophic Forgetting
FRPO applies a max-min robust optimization over KL-bounded policy neighborhoods during RLHF to reduce catastrophic forgetting of safety and accuracy under subsequent SFT or RL fine-tuning.
-
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.
-
Optimizing Chlorination in Water Distribution Systems via Surrogate-assisted Neuroevolution
Surrogate-assisted neuroevolution produces Pareto-optimal chlorine dosing policies for water distribution systems that outperform PPO on four practical objectives.
-
The hidden risks of temporal resampling in clinical reinforcement learning
Resampling clinical time series into uniform bins for offline RL reduces performance by up to 60% and causes retrospective evaluations to overestimate returns by 1.5-3x versus unprocessed data.
-
Diffusion-State Policy Optimization for Masked Diffusion Language Models
DiSPO optimizes intermediate decisions in masked diffusion LMs by branching at selected masked states, resampling tokens, scoring completions, and updating only new tokens using a derived policy-gradient estimator that reuses terminal rollouts.
-
Adaptive Prompt Elicitation for Text-to-Image Generation
Adaptive Prompt Elicitation (APE) uses an information-theoretic framework to generate visual queries that elicit and compile user intent into better prompts for text-to-image models, showing improved alignment in benchmarks and a user study.
-
Rethinking the Design Space of Reinforcement Learning for Diffusion Models: On the Importance of Likelihood Estimation Beyond Loss Design
An ELBO-based likelihood estimator from the final generated sample dominates other RL design factors for diffusion models, raising GenEval from 0.24 to 0.95 in 90 GPU hours with better efficiency than prior methods.
-
Understanding and Exploiting Weight Update Sparsity for Communication-Efficient Distributed RL
PULSE exploits BF16-invisible sparsity in weight updates to enable over 100x lower communication in distributed RL post-training via compute-visible sparsification.
-
HUSKY: Humanoid Skateboarding System via Physics-Aware Whole-Body Control
HUSKY combines humanoid-skateboard dynamics modeling with adversarial motion priors and physics-guided lean-to-steer strategies to achieve real-world stable skateboarding on a humanoid robot.
-
ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning
ESSAM matches PPO and GRPO accuracy (~78%) on GSM8K math tasks but uses 10-18x less GPU memory and shows stronger generalization across datasets.
-
Toward Reliable Sim-to-Real Predictability for MoE-based Robust Quadrupedal Locomotion
MoE-based locomotion policy with RoboGauge metrics achieves reliable sim-to-real transfer, enabling robust quadrupedal walking on challenging unseen terrains up to 4 m/s.
-
Structure Over Scale: Learning Visual Reasoning from Pedagogical Video
Fine-tuning VLMs on 10K QA pairs from pedagogical children's videos produces consistent gains on NExT-QA, Video-MME, and MotionBench, indicating that explicit structure can substitute for data scale.