marinaretikof

marinaretik's activity

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

upvoted 3 papers 9 days ago

Agent-Editing World Model: Rethinking World Modeling for LLM Agents

Paper • 2609.28416 • Published 13 days ago • 43

Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

Paper • 2609.29845 • Published 12 days ago • 104

Training Object Permanence in World Models

Paper • 2609.28654 • Published 13 days ago • 243

upvoted 2 papers 17 days ago

SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness

Paper • 2609.20519 • Published 19 days ago • 139

Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents

Paper • 2609.17708 • Published 21 days ago • 78

upvoted 15 papers 26 days ago

Code World Model: Coding Agent as World Brain

Paper • 2608.25927 • Published Aug 26 • 40

VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning

Paper • 2608.26105 • Published Aug 26 • 193

Procedura: Agentic 3D Modeling with Procedural Control

Paper • 2608.26238 • Published Aug 26 • 16

WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution

Paper • 2608.27454 • Published Aug 27 • 36

PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents

Paper • 2608.26530 • Published Aug 27 • 36

Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report

Paper • 2608.15763 • Published Aug 22 • 54

What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents

Paper • 2608.27260 • Published Aug 27 • 74

Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher

Paper • 2608.26872 • Published Aug 27 • 65

PAWBench: How Far Are We from Probabilistically Aligned World Modeling?

Paper • 2608.27345 • Published Aug 27 • 77

LMSM: LLM Security Framework Inspired by Linux Security Modules

Paper • 2608.25697 • Published Aug 26 • 6

J-Zero: Unified Challenger--Solver--Judge Co-Evolution from Zero Data

Paper • 2608.26582 • Published Aug 27 • 41

Code as Worlds: Agentic Discovery of Executable World Representations for Physical Reasoning

Paper • 2608.27549 • Published Aug 27 • 49

ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models

Paper • 2608.29847 • Published Aug 30 • 6

DreamX-Creator: Democratizing Native Audio-Video Generation at 2K Resolution

Paper • 2608.31106 • Published Aug 31 • 96

Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement

Paper • 2608.31046 • Published Aug 31 • 97