rank-and-files(Ben Rank)

• rank-and-files 27 May 2024 14:35 UTC

  2 points

  1

  on: Train­ing-time do­main au­tho­riza­tion could be helpful for safety

  I agree that this is a very important area of research. In fact, I work on this problem myself.

  Some points:

  1. I didn’t get from the paper alone what $I$ refers to. Maybe a quick definition in the paper would be nice.

  2. I think it would be good to compare against the Vaccine algorithm from Huang et al. (“Vaccine: Perturbation-aware alignment for large language model”) since they are essentially trying to solve the same problem. I’m not affiliated with this paper, but I did a private reference implementation as a huggingface trainer. Lmk if you are interested and I can send you the code.

  3. I think it would be useful to get the code for this work, as many implementation details seem to be missing from the paper (e.g. on my skim I didn’t find the batch-size which you used for training). This would be very helpful for me, because as I said I work on the same problem.