Search papers, labs, and topics across Lattice.
3
8
8
13
Gemma 4's unified architecture and reasoning mode enable it to outperform larger models in human-rated tasks while maintaining high efficiency.
Bounded context windows in next-token prediction models can be fundamentally incompatible with low adversarial regret, even with long context lengths.
Pokemon, not just a childhood game, emerges as a surprisingly effective benchmark for AI, revealing critical gaps in LLMs and RL agents that existing benchmarks miss.
