LAMEN Transcripts

홈페이지 · 논문 1편

Raw negotiation transcripts generated for the paper "Evaluating Language Model Agency through Negotiations". The data includes transcripts from self-play (a model plays against an independent version of itself; corresponding to Section 4.1 of the paper) and cross-play (a model plays against another model; Section 4.2). This dataset encompasses 2926 transcripts (942 self-play, 1984 cross-play).