| ▲ | kgeist a day ago |
| Probably an instance of: "The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A" https://arxiv.org/abs/2309.12288 |
|
| ▲ | hasteg a day ago | parent | next [-] |
| Really interesting paper, thanks for the share. The point their making in that paper reminds me of this paper some people shared around work earlier this year, https://arxiv.org/pdf/2512.14982 (Prompt Repetition Improves Non-Reasoning LLMs)... I wonder how OPs question would fare (or the questions presented in the paper you posted) given double repetition. |
| |
| ▲ | fc417fc802 a day ago | parent [-] | | A relevant question for the paper you linked is what happens if instead of repeating the input prompt you repeat a filler character the same number of times. | | |
| ▲ | matankal 8 hours ago | parent | next [-] | | This is tested in the paper (see the "padding" ablation) | |
| ▲ | hasteg a day ago | parent | prev [-] | | Can you elaborate on that more? Why just a filler character? | | |
| ▲ | fc417fc802 a day ago | parent [-] | | Because padding the output in such a manner (which keep in mind is immediately fed back as input) has been shown to increase model performance. It's one of the many reasons to question what "thinking" traces are really doing. If you consider how the attention mechanism works then a very hand wavey intuition is that despite being entirely arbitrary additional tokens should still provide the opportunity for additional information processing. |
|
|
|
|
| ▲ | _jackdk_ a day ago | parent | prev | next [-] |
| This is also something to be aware of when teaching people, too. I've seen advice for designing Anki-style flashcard decks that reminds people to create flashcards for both A->B and B->A. |
|
| ▲ | hahahaa a day ago | parent | prev [-] |
| Can an LLM self-fine tune by generating reversals? |
