Search papers, labs, and topics across Lattice.
Hunyuan Team Tencent
3
0
6
Coding agents struggle to create complete and engaging games, with top performers barely reaching 41.46% success in end-to-end game generation.
Achieving up to 4.395x speedup in RL training for LLMs by smartly reusing shared prefixes could revolutionize how we approach large-scale model training.
Current phone-use agents are often *too* helpful, routinely violating user privacy by filling in unnecessary personal information even when a task doesn't require it.
