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%).
1926
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.
IRumAI is the first RL agent for Indian Rummy, trained on weak heuristics to beat strong search opponents at 7000x speed.
DeFAb is a large-scale, formally verifiable benchmark for defeasible abduction derived from 18 knowledge bases, demonstrating that frontier LLMs achieve 7.8-65% accuracy versus 100% for a rule-based solver with polynomial-time checks.
Machine learning discovers a tube-seeding strategy for IBP reduction of Feynman integrals that scales linearly with numerator power, demonstrated on rank-20 2-loop 5-point integrals.
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.
citing papers explorer
-
Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems
Mult-DPO provides a tractable multinomial surrogate for set-wise preference optimization in DPO that upper bounds the Plackett-Luce based loss for LLM recommender systems.
-
Beyond One-Size-Fits-All: Adaptive Test-Time Augmentation for Sequential Recommendation
AdaTTA is an actor-critic RL framework that selects sequence-specific test-time augmentations and improves recommendation metrics by up to 26% over fixed augmentation strategies on four datasets.
-
Retrieval Augmented Conversational Recommendation with Reinforcement Learning
RAR retrieves candidate items from a 300k-movie corpus then uses LLM generation with RL feedback to produce context-aware recommendations that outperform baselines on benchmarks.
-
Trustworthiness in Retrieval-Augmented Generation Systems: A Survey
Introduces Trust-RAG Compass framework and TRC Bench benchmark to assess RAG trustworthiness across factuality, robustness, fairness, transparency, accountability, and privacy, with evaluations showing performance gaps between LLMs.
-
Diffusion-GR2: Diffusion Generative Reasoning Re-ranker
Diffusion-GR2 converts an AR reasoning re-ranker to block-diffusion via CFT, OPD, and RL stages, recovering near-parity accuracy on Amazon Beauty with 2.4-3.5x decode speedup.
-
Fairness Attacks on Recommender Systems
A structure-aware RL fairness attack with joint item and gender selection policies is introduced and shown effective on four recommender models across two datasets.
-
Graph-GRPO: Dependency-Aware Credit Assignment for Generative E-commerce Search Relevance
Graph-GRPO builds a dependency graph over CoT steps and propagates outcome rewards to enable finer credit assignment in generative relevance modeling for e-commerce search.
-
Credit-assigned Policy Gradient for Early Stage Retrieval in Two-stage Ranking
CA-PG reduces variance in Plackett-Luce ESR training by computing gradients on marginal item-inclusion probabilities rather than joint candidate-set probabilities.
-
Bridging Textual Profiles and Latent User Embeddings for Personalization
BLUE aligns LLM-generated textual user profiles with embedding-based recommendation objectives via reinforcement learning and next-item text supervision, yielding better zero-shot performance and cross-domain transfer than baselines.
-
Scattered Hypothesis Generation for Open-Ended Event Forecasting
SCATTER uses RL with a hybrid reward combining validity, intra-group diversity, and inter-group diversity to produce inclusive hypothesis sets for event forecasting and outperforms baselines on OpenForecast and OpenEP.
-
Self-Distilled Reinforcement Learning for Co-Evolving Agentic Recommender Systems
CoARS enables co-evolving recommender and user agents by using interaction-derived rewards and self-distilled credit assignment to internalize multi-turn feedback into model parameters, outperforming prior agentic baselines.
-
ReRec: Reasoning-Augmented LLM-based Recommendation Assistant via Reinforcement Fine-tuning
ReRec uses reinforcement fine-tuning with dual-graph reward shaping, reasoning-aware advantage estimation, and online curriculum scheduling to improve LLM reasoning and performance in recommendation tasks.
-
LLM-Enhanced Reinforcement Learning for Long-Term User Satisfaction in Interactive Recommendation
LERL combines an LLM high-level planner for diverse semantic categories with a low-level RL policy for item selection to improve long-term user satisfaction and reduce content homogeneity in interactive recommenders.
-
Learning to Trust: Dynamic Utilization of Retrieval-Augmented Generation for E-commerce Search Relevance
DyKnow-RAG uses Group Relative Policy Optimization with dual-group rollouts and posterior-driven advantage scaling to optimize context utilization in RAG for e-commerce relevance, showing offline gains and production lifts when deployed at Taobao.
-
Learning User-Aware Recall: Personalized Retrieval in Long-Term Conversational Memory
PPRO improves user-aware memory retrieval in conversational agents by using derived user profiles for ranking and training a query rewriter via Group Relative Policy Optimization, with reported gains on LoCoMo and LongMemEval-S benchmarks.
-
TPMM-DPO: Trajectory-aware Preference-guided Model Merging for Iterative Direct Preference Optimization
TPMM-DPO applies trajectory-aware learned-weight merging of prior policy models to stabilize iterative DPO against preference noise accumulation.
-
Efficient Generative Retrieval for E-commerce Search with Semantic Cluster IDs and Expert-Guided RL
CQ-SID semantic IDs and EG-GRPO RL improve generative retrieval hit rates up to 26.76% over RQ-VAE baselines and deliver +1.15% GMV in live e-commerce A/B tests.
-
BatchBench: Toward a Workload-Aware Benchmark for Autoscaling Policies in Big Data Batch Processing -- A Proposed Framework
BatchBench is a proposed framework with workload taxonomy, parameterized generator, five-axis evaluation harness, and standardized agent interface to enable fair comparison of autoscaling policies.
-
UserGPT Technical Report
UserGPT introduces a generative LLM framework with a behavior simulation engine, semantization module, and DF-GRPO post-training that scores 0.7325 on tag prediction and 0.7528 on summary generation on HPR-Bench while compressing records by up to 97.9%.
-
SUMMIR: A Hallucination-Aware Framework for Ranking Sports Insights from LLMs
SUMMIR is a multimetric ranking model that orders LLM-generated sports insights by importance while incorporating hallucination detection to improve factual reliability across cricket, soccer, basketball, and baseball articles.
-
Multimodal Large Language Models with Adaptive Preference Optimization for Sequential Recommendation
HaNoRec dynamically weights harder preference samples and applies Gaussian perturbations to output distributions to improve multimodal LLM performance on sequential recommendation tasks.
-
OneRec-V2 Technical Report
OneRec-V2 scales generative recommendation to 8B parameters via decoder-only design and real-world preference alignment, improving user engagement metrics in production A/B tests.
-
OPERA: A Reinforcement Learning--Enhanced Orchestrated Planner-Executor Architecture for Reasoning-Oriented Multi-Hop Retrieval
OPERA is a new orchestrated planner-executor framework with MAPGRPO training that claims superior results on reasoning-oriented multi-hop retrieval benchmarks.
-
Curr-RLCER:Curriculum Reinforcement Learning For Coherence Explainable Recommendation
Curr-RLCER applies curriculum reinforcement learning with coherence-driven rewards to align generated explanations with predicted ratings in explainable recommendation systems.
-
When & How to Write for Personalized Demand-aware Query Rewriting in Video Search
WeWrite mines user logs to decide when personalization is needed and trains LLMs with SFT and GRPO to rewrite video search queries, delivering 1.07% more long-view clicks and 2.97% fewer reformulations in live A/B tests.