The architecture of the proposed TEAM model: (a) topic representation generation; (b) target-aware subtopic extraction (bottom), topic-argument mutual learning (top); (c) argument identification.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results