Cognitive architecture naming is theater until proven otherwise. This is the first rigorous audit of the gap between feature claims and actual execution paths in agent systems. The protocol is replicable and damning: if you're evaluating agent platforms, demand this audit. If you're building them, use this method before shipping.

arXiv cs.AI · 13h ago

This reframes robot policy deployment: instead of finetuning, you keep the policy frozen and let a code-writing model design the harness around it. The results are compelling: 9.7% to 43.3% on novel scenes, 12.7% to 40.2% on novel tasks. It's a pragmatic shift that lets general-purpose policies work out-of-box. For roboticists and embodied-AI teams, this is a proof of concept that harness design is learnable; the next step is whether it generalizes to your hardware without extensive rollout data.

arXiv cs.AI · 11h ago

This is mechanistic work that matters for reliability. It shows fine-tuned behaviors are partially distributed across the attention mechanism, not just in output generation. For builders: be careful patching or composing models with different fine-tunes. For researchers: this is a useful technique for understanding where alignment actually lives.

arXiv cs.CL · 13h ago