{"total":72,"rows":[{"id":"2601.02780","title":"MiMo-V2-Flash Technical Report","created":"2026-01-06","cited_by_count":191,"area":"moe"},{"id":"2512.01374","title":"Stabilizing Reinforcement Learning with LLMs: Formulation and Practices","created":"2025-12-01","cited_by_count":41,"area":"moe"},{"id":"2602.04879","title":"Rethinking the Trust Region in LLM Reinforcement Learning","created":"2026-02-04","cited_by_count":37,"area":"lr"},{"id":"2603.25040","title":"Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale","created":"2026-03-26","cited_by_count":29,"area":null},{"id":"2603.24477","title":"Composer 2 Technical Report","created":"2026-03-25","cited_by_count":22,"area":"agents"},{"id":"2603.07685","title":"Scalable Training of Mixture-of-Experts Models with Megatron Core","created":"2026-03-08","cited_by_count":19,"area":"moe"},{"id":"2602.20739","title":"PyVision-RL: Forging Open Agentic Vision Models via RL","created":"2026-02-24","cited_by_count":16,"area":"agents"},{"id":"2601.18150","title":"FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning","created":"2026-01-26","cited_by_count":13,"area":"inference"},{"id":"2602.04705","title":"ERNIE 5.0 Technical Report","created":"2026-02-04","cited_by_count":11,"area":"moe"},{"id":"2602.10693","title":"VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training","created":"2026-02-11","cited_by_count":11,"area":"moe"},{"id":"2602.17616","title":"Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs","created":"2026-02-19","cited_by_count":11,"area":null},{"id":"2603.01875","title":"KDFlow: A User-Friendly and Efficient Knowledge Distillation Framework for Large Language Models","created":"2026-03-02","cited_by_count":7,"area":"distill"},{"id":"2604.11554","title":"Relax: An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale","created":"2026-04-13","cited_by_count":7,"area":"lr"},{"id":"2605.14220","title":"Diagnosing Training Inference Mismatch in LLM Reinforcement Learning via a Zero-Mismatch Reference","created":"2026-05-14","cited_by_count":7,"area":null},{"id":"2609.11042","title":"T1: Terminal Agent Reinforcement Learning for Long-Horizon Tasks","created":"2026-09-10","cited_by_count":7,"area":"lr"},{"id":"2605.19577","title":"GoLongRL: Capability-Oriented Long Context Reinforcement Learning with Multitask Alignment","created":"2026-05-19","cited_by_count":6,"area":null},{"id":"2606.09348","title":"PBSD: Privileged Bayesian Self-Distillation for Long-Horizon Credit Assignment","created":"2026-06-08","cited_by_count":6,"area":"lr"},{"id":"2603.06626","title":"Grouter: Decoupling Routing from Representation for Accelerated MoE Training","created":"2026-02-22","cited_by_count":5,"area":"moe"},{"id":"2603.10395","title":"Graph-GRPO: Training Graph Flow Models with Reinforcement Learning","created":"2026-03-11","cited_by_count":5,"area":"lr"},{"id":"2605.12070","title":"Missing Old Logits in Asynchronous Agentic RL: Semantic Mismatch and Repair Methods for Off-Policy Correction","created":"2026-05-12","cited_by_count":5,"area":"agents"},{"id":"2607.18722","title":"Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning","created":"2026-07-21","cited_by_count":5,"area":"moe"},{"id":"2607.22925","title":"Not All LLM Reasoning is Visible in the Chain-of-Thought","created":"2026-07-24","cited_by_count":5,"area":"lr"},{"id":"2608.08888","title":"Full-bandwidth transformer","created":"2026-08-09","cited_by_count":5,"area":null},{"id":"2605.05365","title":"ZAYA1-8B Technical Report","created":"2026-05-06","cited_by_count":4,"area":"lr"},{"id":"2608.13505","title":"Intern-S2-Preview: Scientific Agentic Foundation Model","created":"2026-08-13","cited_by_count":4,"area":"agents"},{"id":"2608.17393","title":"LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents","created":"2026-08-18","cited_by_count":4,"area":"lr"},{"id":"2601.06052","title":"Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization","created":"2025-12-19","cited_by_count":3,"area":"lr"},{"id":"2602.01103","title":"Probing RLVR training instability through the lens of objective-level hacking","created":"2026-02-01","cited_by_count":3,"area":"lr"},{"id":"2605.04077","title":"Balanced Aggregation: Understanding and Fixing Aggregation Bias in GRPO","created":"2026-04-14","cited_by_count":3,"area":"lr"},{"id":"2605.08639","title":"ReLibra: Routing-Replay-Guided Load Balancing for MoE Training in Reinforcement Learning","created":"2026-05-09","cited_by_count":3,"area":"moe"},{"id":"2606.03070","title":"ASymPO: Asymmetric-Scale Policy Optimization for Asynchronous LLM Post-Training Without Behavior Information","created":"2026-06-02","cited_by_count":3,"area":null},{"id":"2609.20807","title":"Score Centering Stabilizes Off-policy Reinforcement Learning","created":"2026-09-17","cited_by_count":3,"area":"lr"},{"id":"2511.04285","title":"RLoop: An Self-Improving Framework for Reinforcement Learning with Iterative Policy Initialization","created":"2025-11-06","cited_by_count":2,"area":"lr"},{"id":"2512.07710","title":"Each Prompt Matters: Scaling Reinforcement Learning Without Wasting Rollouts on Hundred-Billion-Scale MoE","created":"2025-12-08","cited_by_count":2,"area":"lr"},{"id":"2605.06326","title":"Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning","created":"2026-05-07","cited_by_count":2,"area":"lr"},{"id":"2605.29697","title":"Beyond Trajectory Rewards: Step-level Credit Assignment for Agentic Search via Graph Modeling","created":"2026-05-28","cited_by_count":2,"area":"agents"},{"id":"2606.11867","title":"Harnessing Routing Foresight for Micro-step-level MoE load balancing in RL Post-training","created":"2026-06-10","cited_by_count":2,"area":"moe"},{"id":"2607.04728","title":"Turning Off-Policy Tokens On-Policy: A Plug-in Approach for Improving LLM Alignment","created":"2026-07-06","cited_by_count":2,"area":"alignment"},{"id":"2608.09251","title":"MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts","created":"2026-08-10","cited_by_count":2,"area":"agents"},{"id":"2609.04128","title":"Environment Evolution for Terminal Agents","created":"2026-09-03","cited_by_count":2,"area":"agents"},{"id":"2605.13779","title":"MinT: Managed Infrastructure for Training and Serving Millions of LLMs","created":"2026-05-13","cited_by_count":1,"area":"moe"},{"id":"2606.00395","title":"PR2: Predictive Routing Replay for MoE-Based LLM Reinforcement Learning","created":"2026-05-29","cited_by_count":1,"area":"moe"},{"id":"2606.26607","title":"Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch","created":"2026-06-25","cited_by_count":1,"area":"moe"},{"id":"2608.09819","title":"Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA","created":"2026-08-10","cited_by_count":1,"area":null},{"id":"2609.08368","title":"Miles v0.1: Production-Level Post-Training","created":"2026-09-08","cited_by_count":1,"area":null},{"id":"2610.11959","title":"MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement","created":"2026-10-08","cited_by_count":1,"area":"lr"},{"id":"2601.22532","title":"Demystifying Design Choices of Reinforcement Fine-tuning: A Batched Contextual Bandit Learning Perspective","created":"2026-01-30","cited_by_count":0,"area":null},{"id":"2604.03342","title":"Mixture-of-Experts in Remote Sensing: A Survey","created":"2026-04-03","cited_by_count":0,"area":"moe"},{"id":"2605.12380","title":"Trust the Batch, On- or Off-Policy: Adaptive Policy Optimization for RL Post-Training","created":"2026-05-12","cited_by_count":0,"area":"deeprl"},{"id":"2605.28109","title":"Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization","created":"2026-05-27","cited_by_count":0,"area":null},{"id":"2606.10369","title":"PADD: Path-Aligned Decompression Distillation for Non-Router Teacher to Guide MoE Student Learning","created":"2026-06-09","cited_by_count":0,"area":"moe"},{"id":"2606.16154","title":"A Gradient Perspective on RLVR Stability and Winner Advantage Policy Optimization","created":"2026-06-15","cited_by_count":0,"area":"lr"},{"id":"2606.24901","title":"LLM Evolution as an Industry-Scale Ecosystem: A Lifecycle Perspective on Continual Learning","created":"2026-06-12","cited_by_count":0,"area":null},{"id":"2607.06987","title":"UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma","created":"2026-07-08","cited_by_count":0,"area":"lr"},{"id":"2607.21653","title":"Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement Learning","created":"2026-07-22","cited_by_count":0,"area":"agents"},{"id":"2607.22186","title":"Deconstructing Off-Policy Ratios: Entropy-Normalized Trust Regions for Asynchronous Reinforcement Learning","created":"2026-07-24","cited_by_count":0,"area":"deeprl"},{"id":"2607.23124","title":"AgentOmnia: Scaling Agentic Models for Full-Scenario Applications","created":"2026-07-25","cited_by_count":0,"area":"agents"},{"id":"2608.12146","title":"RoutePack: Expert Placement and Attention-Aware Data Packing for MoE Reinforcement Learning","created":"2026-08-12","cited_by_count":0,"area":"moe"},{"id":"2608.15787","title":"Routing Divergence Is Not Evidence of Behavioral Influence in Same-Weight MoE Self-Distillation","created":"2026-08-16","cited_by_count":0,"area":"moe"},{"id":"2609.00791","title":"Instella-MoE Technical Report","created":"2026-09-01","cited_by_count":0,"area":"moe"},{"id":"2609.09783","title":"BRACE: Anchored Bellman-Residual Correction for Stale Critics in Asynchronous RL","created":"2026-09-09","cited_by_count":0,"area":null},{"id":"2609.13058","title":"Expert-Space Exploration in MoE Reinforcement Learning","created":"2026-09-11","cited_by_count":0,"area":"moe"},{"id":"2609.15987","title":"Bellman Policy Optimization","created":"2026-09-14","cited_by_count":0,"area":"lr"},{"id":"2609.29142","title":"Not Every Token Is Worth Distilling: Selective Supervision for Direct-OPD","created":"2026-09-24","cited_by_count":0,"area":"distill"},{"id":"2609.29421","title":"Rufus-Air: An Open LLM Post-Training Recipe","created":"2026-09-24","cited_by_count":0,"area":"alignment"},{"id":"2609.32444","title":"Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It","created":"2026-09-26","cited_by_count":0,"area":"lr"},{"id":"2609.35433","title":"ReSPO: Reshaped Sequence Policy Optimization for Gradient Starvation in Off-Policy Learning","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2610.05458","title":"Measuring and Reducing Cross-Vendor Mismatch in Language Models","created":"2026-10-04","cited_by_count":0,"area":"moe"},{"id":"2610.09914","title":"RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement Learning","created":"2026-10-07","cited_by_count":0,"area":"lr"},{"id":"2610.16352","title":"A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated Recipes","created":"2026-10-12","cited_by_count":0,"area":"agents"},{"id":"2610.18010","title":"Fish Audio Drama 3 Technical Report","created":"2026-10-13","cited_by_count":0,"area":"moe"},{"id":"2610.20538","title":"PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis","created":"2026-10-14","cited_by_count":0,"area":null}],"next_cursor":null}