Search papers, labs, and topics across Lattice.
UC San Diego
3
0
3
Self-looping reasoning in LLMs can be effectively mitigated through targeted hidden-state interventions, leading to improved reasoning quality and efficiency.
Achieving near-state-of-the-art performance without training, ZoRRO revolutionizes the speed and efficiency of personalized news recommendations.
NAILS aligns recommender systems with normative objectives without retraining, balancing user engagement with fairness and diversity.
