Search papers, labs, and topics across Lattice.
Affiliation:
3
7
7
5
Self-looping reasoning in LLMs can be effectively mitigated through targeted hidden-state interventions, leading to improved reasoning quality and efficiency.
Rollout design in LLM reinforcement learning is more than just sampling trajectories – it's a modular pipeline you can optimize for reliability, coverage, and cost.
Synthetic data from non-conversational signals can outperform real dialogue in low-resource Conversational Recommender Systems, transforming how we approach data scarcity.
