Search papers, labs, and topics across Lattice.
3
0
4
Cross-lingual exploration can unlock hidden knowledge in LLMs, improving factual recall and consistency across 17 languages.
Adversarial training can significantly enhance the robustness of parameter-efficient fine-tuning, especially in low-resource NLP scenarios.
Average reward RL can finally handle the messy reality of non-stationary rewards and durations in SMDPs, thanks to a clever harmonic mean trick.
