Showing 2024Show all
3 papers · 1 filter
cs.LG2024
Towards a Theoretical Understanding of the 'Reversal Curse' via Training Dynamics
Hanlin Zhu, Baihe Huang, Shaolun Zhang +4
Auto-regressive large language models (LLMs) show impressive capacities to solve many complex reasoning tasks while struggling with some simple logical reasoning tasks such as inve…
cs.CL2024
Efficient Prompt Caching via Embedding Similarity
Hanlin Zhu, Banghua Zhu, Jiantao Jiao
Large language models (LLMs) have achieved huge success in numerous natural language process (NLP) tasks. However, it faces the challenge of significant resource consumption during…
cs.LG2024
Avoiding Catastrophe in Online Learning by Asking for Help
Benjamin Plaut, Hanlin Zhu, Stuart Russell
Most learning algorithms with formal regret guarantees assume that all mistakes are recoverable and essentially rely on trying all possible behaviors. This approach is problematic…