HINT-WAM: Hindsight-Informed Training for World Action Models Paper • 2611.24789 • Published 5 days ago • 23
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published Sep 29 • 146
VisionHOPE: Visual Backbones as Self-Modifying Learning Systems Paper • 2609.33325 • Published Sep 27 • 53
GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis Paper • 2609.38923 • Published Sep 30 • 58
SemaPLC: A Project-Grounded, Verification-Gated Agent Harness for PLC Code Generation Paper • 2608.18565 • Published Aug 19 • 20