MIT researchers developed Attention Matching, a KV cache compaction technique that compresses LLM memory by 50x in seconds — ...

Claude Code Skills 2.0 adds evals plus benchmark test sets; changes target skill reliability as models update over time.

Quantum computing advantages look weaker; classical methods beat a nitrogen-fixing molecule simulation, raising doubts about ...

Magnetic invisibility sounds simple in theory. Place the right materials around an object and magnetic fields flow around it as if nothing were there. Reality has been far messier. For nearly two ...

This research advances hybrid soft-rigid robot simulations, achieving up to 1000 times faster computations through analytical derivatives in the GVS framework.