{"total":66,"rows":[{"id":"2601.22060","title":"Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models","created":"2026-01-29","cited_by_count":37,"area":"mllm"},{"id":"2602.04879","title":"Rethinking the Trust Region in LLM Reinforcement Learning","created":"2026-02-04","cited_by_count":37,"area":"lr"},{"id":"2511.06411","title":"SofT-GRPO: Surpassing Discrete-Token LLM Reinforcement Learning via Gumbel-Reparameterized Soft-Thinking Policy Optimization","created":"2025-11-09","cited_by_count":21,"area":"lr"},{"id":"2604.06268","title":"RAGEN-2: Reasoning Collapse in Agentic RL","created":"2026-04-07","cited_by_count":20,"area":"lr"},{"id":"2511.18538","title":"From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence","created":"2025-11-23","cited_by_count":15,"area":"agents"},{"id":"2511.17826","title":"Deterministic Inference across Tensor Parallel Sizes That Eliminates Training-Inference Mismatch","created":"2025-11-21","cited_by_count":14,"area":null},{"id":"2601.10201","title":"Future-KL Regularized GRPO: Process-Level Credit Assignment from $f$-Divergence Regularization","created":"2026-01-15","cited_by_count":12,"area":"lr"},{"id":"2602.17616","title":"Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs","created":"2026-02-19","cited_by_count":11,"area":null},{"id":"2603.11682","title":"Entropy-Preserving Reinforcement Learning","created":"2026-03-12","cited_by_count":11,"area":"lr"},{"id":"2604.04917","title":"Vero: An Open RL Recipe for General Visual Reasoning","created":"2026-04-06","cited_by_count":11,"area":"lr"},{"id":"2512.24618","title":"Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models","created":"2025-12-31","cited_by_count":10,"area":"agents"},{"id":"2602.19362","title":"LLMs Can Learn to Reason Via Off-Policy RL","created":"2026-02-22","cited_by_count":10,"area":"lr"},{"id":"2602.01826","title":"Beyond Precision: Training-Inference Mismatch is an Optimization Problem and Simple LR Scheduling Fixes It","created":"2026-02-02","cited_by_count":9,"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":"2604.07853","title":"QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training--Inference Mismatch","created":"2026-04-09","cited_by_count":7,"area":null},{"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":"2605.17570","title":"How Off-Policy Can GRPO Be? Mu-GRPO for Efficient LLM Reinforcement Learning","created":"2026-05-17","cited_by_count":7,"area":"lr"},{"id":"2602.12036","title":"Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models","created":"2026-02-12","cited_by_count":6,"area":null},{"id":"2604.21199","title":"ARFBench: Benchmarking Time Series Question Answering Ability for Software Incident Response","created":"2026-04-23","cited_by_count":6,"area":null},{"id":"2605.29398","title":"GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models","created":"2026-05-28","cited_by_count":6,"area":"diffusion"},{"id":"2604.19790","title":"Hidden Reliability Risks in Large Language Models: Systematic Identification of Precision-Induced Output Disagreements","created":"2026-04-02","cited_by_count":5,"area":null},{"id":"2606.08446","title":"Sparrow: Sparse Rollout for Stable and Efficient Long-context RL of Large Language Models","created":"2026-06-07","cited_by_count":5,"area":"lr"},{"id":"2606.09821","title":"Rethinking the Divergence Regularization in LLM RL","created":"2026-06-08","cited_by_count":5,"area":"lr"},{"id":"2512.16144","title":"INTELLECT-3: Technical Report","created":"2025-12-18","cited_by_count":4,"area":"lr"},{"id":"2601.15141","title":"CLEANER: Self-Purified Trajectories Boost Agentic Reinforcement Learning","created":"2026-01-21","cited_by_count":4,"area":"agents"},{"id":"2602.09591","title":"On the Optimal Reasoning Length for RL-Trained Language Models","created":"2026-02-10","cited_by_count":4,"area":"lr"},{"id":"2605.05365","title":"ZAYA1-8B Technical Report","created":"2026-05-06","cited_by_count":4,"area":"lr"},{"id":"2606.08094","title":"vla.cpp: A Unified Inference Runtime for Vision-Language-Action Models","created":"2026-06-06","cited_by_count":4,"area":"vla"},{"id":"2606.12594","title":"Pythagoras-Prover: Advancing Efficient Formal Proving via Augmented Lean Formalisation","created":"2026-06-10","cited_by_count":4,"area":"lr"},{"id":"2608.23566","title":"Best Practice Critic Optimization","created":"2026-08-24","cited_by_count":4,"area":"lr"},{"id":"2512.05962","title":"Whatever Remains Must Be True: Filtering Drives Reasoning in LLMs, Shaping Diversity","created":"2025-12-05","cited_by_count":3,"area":"lr"},{"id":"2603.19470","title":"Adaptive Layerwise Perturbation: Unifying Off-Policy Corrections for LLM RL","created":"2026-03-19","cited_by_count":3,"area":null},{"id":"2607.15810","title":"QUADS: Stabilizing NVFP4 Reinforcement Learning for MoE via QUantization-error Alignment across Dual Sides","created":"2026-07-17","cited_by_count":3,"area":"moe"},{"id":"2609.20807","title":"Score Centering Stabilizes Off-policy Reinforcement Learning","created":"2026-09-17","cited_by_count":3,"area":"lr"},{"id":"2511.02826","title":"PLUTO-4: Frontier Pathology Foundation Models","created":"2025-11-04","cited_by_count":2,"area":null},{"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":"2605.08862","title":"BubbleSpec: Turning Long-Tail Bubbles into Speculative Rollout Drafts for Synchronous Reinforcement Learning","created":"2026-05-09","cited_by_count":2,"area":"lr"},{"id":"2606.21023","title":"Demystifying Numerical Instability in LLM Inference: Achieving Reproducible Inference for Mission-Critical Tasks with HEAL","created":"2026-06-19","cited_by_count":2,"area":"inference"},{"id":"2512.06392","title":"RLAX: Large-Scale, Distributed Reinforcement Learning for Large Language Models on TPUs","created":"2025-12-06","cited_by_count":1,"area":"lr"},{"id":"2601.14716","title":"PCL-Reasoner-V1.5: Advancing Math Reasoning with Offline Reinforcement Learning","created":"2026-01-21","cited_by_count":1,"area":"lr"},{"id":"2602.03190","title":"PrAg-PO: Prompt Augmented Policy Optimization for Robust and Diverse Mathematical Reasoning","created":"2026-02-03","cited_by_count":1,"area":"lr"},{"id":"2602.10609","title":"Online Causal Kalman Filtering for Stable and Effective Policy Optimization","created":"2026-02-11","cited_by_count":1,"area":null},{"id":"2604.28020","title":"Cost-Aware Learning","created":"2026-04-30","cited_by_count":1,"area":null},{"id":"2605.20865","title":"Multi-Step Likelihood-Ratio Correction for Reinforcement Learning with Verifiable Rewards","created":"2026-05-20","cited_by_count":1,"area":"lr"},{"id":"2606.15621","title":"Re-feeding Is Not Replaying: Measuring Replay Noise in Counterfactual Token-Credit Estimation","created":"2026-06-14","cited_by_count":1,"area":null},{"id":"2606.29526","title":"The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning","created":"2026-06-28","cited_by_count":1,"area":null},{"id":"2607.24062","title":"ACRL: Adaptive Control of Training-Inference Discrepancy for Stable Reinforcement Learning","created":"2026-07-27","cited_by_count":1,"area":"lr"},{"id":"2608.11797","title":"Reading the Gate, Not the Interference: Output-Side Interference Measurement Does Not Track Merge Collapse","created":"2026-08-12","cited_by_count":1,"area":null},{"id":"2602.14134","title":"DenseMLLM: Standard Multimodal LLMs for Dense Prediction","created":"2026-02-15","cited_by_count":0,"area":"mllm"},{"id":"2602.19020","title":"Learning to Detect Language Model Training Data via Active Reconstruction","created":"2026-02-22","cited_by_count":0,"area":"safety"},{"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":"2606.08779","title":"Reformulate LLM Reinforcement Learning for Efficient Training under Black-box Discrepancy","created":"2026-06-07","cited_by_count":0,"area":null},{"id":"2607.10848","title":"Predictive Divergence Masks for LLM RL","created":"2026-07-12","cited_by_count":0,"area":null},{"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":"2608.05126","title":"Spoken Function Calling: A New Perspective on Spoken Language Understanding for Large Audio Language Models","created":"2026-08-05","cited_by_count":0,"area":"agents"},{"id":"2608.05600","title":"LC-GRPO: Bridging Train-Inference Gap for Flow-Based GRPO with Langevin Correction","created":"2026-08-06","cited_by_count":0,"area":"lr"},{"id":"2608.10823","title":"MoE Proxy Models for Low-Cost Failure Reproduction and Diagnosis in LLM RL Post-Training","created":"2026-08-11","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":"2609.15987","title":"Bellman Policy Optimization","created":"2026-09-14","cited_by_count":0,"area":"lr"},{"id":"2609.26621","title":"Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference","created":"2026-09-22","cited_by_count":0,"area":null},{"id":"2609.28838","title":"Uncertainty-Gated Exploration Noise Suppresses Task Collapse in Online RL Fine-Tuning of a Flow-Matching Vision-Language-Action Policy","created":"2026-09-23","cited_by_count":0,"area":"vla"},{"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":"2610.05458","title":"Measuring and Reducing Cross-Vendor Mismatch in Language Models","created":"2026-10-04","cited_by_count":0,"area":"moe"},{"id":"2610.08430","title":"NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter Scale","created":"2026-10-06","cited_by_count":0,"area":"agents"},{"id":"2610.13202","title":"S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding","created":"2026-10-09","cited_by_count":0,"area":"inference"},{"id":"2610.18010","title":"Fish Audio Drama 3 Technical Report","created":"2026-10-13","cited_by_count":0,"area":"moe"}],"next_cursor":null}