Search papers, labs, and topics across Lattice.
5
5
7
8
TRACE transforms user corrections into enforceable rules, slashing preference violations from 100% to as low as 2% in critical coding tasks.
LLMs still struggle to apply public policy knowledge in real-world scenarios, even when they can memorize facts and understand concepts.
LLMs can parrot numerical shortcuts, but they fundamentally lack the human-like "number sense" to know when and why those shortcuts actually work.
LLMs can be taught to be dignified peers instead of evasive sycophants, by carefully balancing anti-sycophancy and trustworthiness with empathy and creativity.
The HHH principle needs a serious makeover: this paper proposes a framework for dynamically prioritizing helpfulness, honesty, and harmlessness based on context, offering a more nuanced approach to AI alignment.
