Search papers, labs, and topics across Lattice.
Affiliation:
5
0
7
4
MidTool reveals that dedicated mid-training can significantly enhance LLMs' ability to utilize tools effectively, outperforming traditional post-training approaches.
Complex search behaviors typically requiring explicit controllers can emerge naturally from an agent's reasoning process, leading to substantial performance gains in optimization tasks.
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.
Stop hand-crafting hints for RL agents: HiLL learns to generate adaptive hints that actually improve the agent's performance on the original task, not just the hinted one.
Forget prompt engineering – LSE trains LLMs to self-edit their own contexts at test time, outperforming even GPT-5 and Claude Sonnet 4.5 in Text-to-SQL and question answering.
