A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity.
Andrew Lee, Xiaoyan Bai, Itamar Pres, Martin Wattenberg, Jonathan K. Kummerfeld, Rada Mihalcea
Browse the full ICML paper archive.
Andrew Lee, Xiaoyan Bai, Itamar Pres, Martin Wattenberg, Jonathan K. Kummerfeld, Rada Mihalcea
Browse the full ICML paper archive.