RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations
Paper • 2610.01780 • Published • 285
Themba Nkosi
themba-n
AI & ML interests
None yet
Recent Activity
View all activity
Organizations
None yet
upvoted 4
papers 8 days ago
Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory
Paper • 2610.02521 • Published • 60
LVMT: Video Mask Transformer for Long-term Video Segmentation
Paper • 2609.34895 • Published • 19
HelixWorld: A Real-time Interactive Audio-Visual World Model
Paper • 2609.38123 • Published • 38
upvoted 7
papers 13 days ago
Act First, Reason Later: Accelerating On-Policy Distillation for Multi-Turn Agents via Reference-Conditioned Inverse Dynamics
Paper • 2609.36608 • Published • 8
Where the Model Changes Its Mind: Hindsight-Divergence Localization for Efficient Reinforcement Learning with Verifiable Rewards
Paper • 2609.36864 • Published • 10
Omni-Decision: Evidence-Ledger Planning for Omni-Modal Agents
Paper • 2607.11433 • Published • 31
ANTMAN: Adaptive Need Tracking for Multi-Agent Navigation in Large Information Spaces
Paper • 2609.33326 • Published • 37
WorldAttention: An Efficient Attention Architecture for Interactive Video World Models
Paper • 2609.34606 • Published • 51
Anisotropic Representations Improve Planning in JEPA World Models
Paper • 2609.37441 • Published • 35
Scaling Properties of Same-Family On-Policy Distillation
Paper • 2609.32722 • Published • 326
