Kaito Miyazaki

kaitomiya's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted 3 papers 1 day ago

EVISKILL: Grounding Skill Evolution in Replayable Evidence

Paper • 2610.05030 • Published 6 days ago • 50

Understanding and Enhancing Backdoor Persistency in LLM Agent Post-Training

Paper • 2610.07510 • Published 5 days ago • 12

TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models

Paper • 2610.07767 • Published 4 days ago • 87

upvoted 3 papers 2 days ago

Foundations of Proactive Agents: Principles, Technical Layers, and Proactivity-Gym

Paper • 2609.37267 • Published 11 days ago • 40

LMBuild: Evaluating LLM Agents for Generating Buildable and Functional Structures

Paper • 2610.04292 • Published 7 days ago • 38

What Gradients Add to Text Leakage in Split Language Models, Counted per Token and per Document

Paper • 2610.04128 • Published 8 days ago • 12

upvoted 3 papers 3 days ago

VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks

Paper • 2610.00972 • Published 9 days ago • 59

Source Preference in the Wild: How LLM Agents Favor Items by Source, and How to Reduce It

Paper • 2610.03195 • Published 8 days ago • 46

HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents

Paper • 2610.03574 • Published 8 days ago • 60

upvoted 2 papers 4 days ago

ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization

Paper • 2610.00906 • Published 9 days ago • 83

X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization

Paper • 2609.32993 • Published 14 days ago • 78

upvoted 4 papers 5 days ago

SkillGym: Training Skill-Use Agents with Automatic Verifiable Environment Generation

Paper • 2609.37539 • Published 11 days ago • 12

EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making

Paper • 2609.38334 • Published 9 days ago • 80

AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks

Paper • 2609.38288 • Published 11 days ago • 136

LoopVL: Recurrent Visual Intelligence

Paper • 2609.38426 • Published 11 days ago • 475

upvoted 4 papers 6 days ago

ExpVoyager: Direct Experience Navigation for Dynamic Agent Skill Synthesis

Paper • 2609.32630 • Published 14 days ago • 19

TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces

Paper • 2609.33295 • Published 13 days ago • 72

Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents

Paper • 2609.29892 • Published 16 days ago • 34

MaLiang-Harness: A Programmable Path to Image and Video Generation

Paper • 2609.34309 • Published 12 days ago • 394

upvoted a paper 7 days ago

IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis

Paper • 2609.29444 • Published 16 days ago • 21