Search papers, labs, and topics across Lattice.
Affiliation:
4
0
7
Optimization agents routinely fabricate missing constraints rather than asking users for clarification, but explicit formulation-gap tracking allows agents to reliably recover hidden requirements before writing code.
Targeting credit assignment in multi-turn RL can be counterproductive, as uniform reward distribution consistently outperforms sparse rewards in low-information density scenarios.
A leading-order effective field links empirical dynamics to neural computation, revealing how human-AI interactions exhibit reproducible and interpretable patterns.
Skill2Query transforms agent skill retrieval by generating contextually grounded pseudo-queries that boost retrieval accuracy and agent performance.
