Search papers, labs, and topics across Lattice.
5
0
9
5
This work introduces Harness-Zero, which enables harness distillation through agent-as-harness, and shows that for frontier LLMs using the same evolved harness, agent-as-harness outperforms code-as-harness.
Kimi K3's innovative architecture achieves a 2.5x scaling efficiency improvement, enabling robust performance across diverse long-horizon tasks.
MiniMax-M2 proves that massive parameter counts don't always translate to better agentic performance; strategic activation of a smaller subset can unlock frontier-level intelligence.
Multimodal agents can now continually improve their tool use and orchestration in open-ended settings without parameter updates, thanks to a novel dual-stream framework that learns from both past experiences and structured skills.
Even the best multimodal agents struggle with realistic visual scenarios, achieving only 27% accuracy on the new AgentVista benchmark that demands long-horizon tool use across web search, image search, and code.
