Search papers, labs, and topics across Lattice.
4
0
7
0
Instead of treating emotion-cause extraction as independent pairwise classification, this work formulates it as a global alignment problem between disentangled emotion and cause representations, leading to improved performance.
LLM agents often say one thing, believe another, and do something completely different, especially when interacting with other agents.
Finer-grained abstract semantics can finally deliver on its promise of faster program synthesis, thanks to a new offline "presynthesis" phase that avoids costly online pruning.
LLM agents can achieve 3x faster web search and higher accuracy by dynamically routing between multiple context management strategies.
