Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs
Paper • 2609.26796 • Published • 36
Karthik Iyer
kiyer97
·
AI & ML interests
Reinforcement learning, reward modeling, RLHF, policy optimization, offline RL
Recent Activity
liked a dataset 4 days ago
youinwww/reinforcement_learning upvoted a paper 4 days ago
Towards Full Pipeline FP8 Reinforcement Learning for LLMsOrganizations
None yet
upvoted a paper 3 days ago
upvoted 4
papers 4 days ago
Towards Full Pipeline FP8 Reinforcement Learning for LLMs
Paper • 2609.22870 • Published • 17
ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models
Paper • 2609.13231 • Published • 18
VideoGen-Agent: Reinforcing Video Generation Agents
Paper • 2609.24997 • Published • 71
One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents
Paper • 2609.23377 • Published • 50
upvoted 6
papers 5 days ago
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Paper • 2609.22068 • Published • 136
Grounded Skill Synthesis from Code at Scale for Agentic Intelligence
Paper • 2609.05571 • Published • 119
StudentSim: Training LLM-based Student Simulators
Paper • 2609.01591 • Published • 447
MintAct: A Unified Visual Agent for Digital Environments
Paper • 2609.22083 • Published • 35
BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence
Paper • 2609.20886 • Published • 30
From Pretraining to Proficiency: Real-World Subtask RL for Long-Horizon Manipulation with Minimal Human Intervention
Paper • 2609.21788 • Published • 13
upvoted 4
papers 6 days ago
Region-Level Policy Optimization for Fine-grained MLLM Perception
Paper • 2609.19745 • Published • 41
Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL
Paper • 2609.20715 • Published • 44
RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning
Paper • 2609.20784 • Published • 57
Verifiable Social Reasoning for LLM Assistants
Paper • 2609.17496 • Published • 67
upvoted 3
papers 8 days ago
Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening
Paper • 2609.18708 • Published • 82
ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments
Paper • 2609.19134 • Published • 101
ActionPiece: Rethinking Action Tokenization for Autoregressive Vision-Language-Action Models
Paper • 2609.18487 • Published • 47
upvoted 2
papers 10 days ago