Can LLMs learn from a single example?
1–10 of 143 posts
Re: Can LLMs learn from a single example?
#2Thank you for posting this to HN! :D
I'm one of the authors of this post -- Johno & I found it really interesting looking into this curious issue of rapid memorization from LLMs. I've been working with neural nets for 30 years, and fine-tuning language models since 2017, and this behavior is most surprising to me! Other folks have seen it in LLMs too, although I haven't seen a analysis of this kind before (although we might have missed something).
Let me know if you have any questions or thoughts.
Re: Can LLMs learn from a single example?
#3Thank you for posting this to HN! :D I'm one of the authors of this post -- Johno & I found it really interesting looking into this curious issue of rapid memorization from LLMs. I've been working with neural nets for 30 years, and fine-tuning language models since 2017, and this behavior is most surprising to me! Other folks have seen it in LLMs too, although I haven't seen a analysis of this kind before (although w…
In the palm-e paper (https://palm-e.github.io/), when they try to unfreeze and train the LLM on new image data only, there is expectedly a lot of CF on NLP tasks but very interestingly, the effect diminishes greatly with the scale of the LLM prior to training.
From an average -87.3% performance drop on the 12B model to -61.6% on the 84B model then just -3.9% on the 562B model. Felt like we were just shy of an insight breakthrough here.
Is avoiding CF potentially just a matter of sheer scale ?
Re: Can LLMs learn from a single example?
#4Thank you for posting this to HN! :D I'm one of the authors of this post -- Johno & I found it really interesting looking into this curious issue of rapid memorization from LLMs. I've been working with neural nets for 30 years, and fine-tuning language models since 2017, and this behavior is most surprising to me! Other folks have seen it in LLMs too, although I haven't seen a analysis of this kind before (although w…
In the palm-e paper ( https://palm-e.github.io/ ), when they try to unfreeze and train the LLM on new image data only, there is expectedly a lot of CF on NLP tasks but very interestingly, the effect diminishes greatly with the scale of the LLM prior to training. From an average -87.3% performance drop on the 12B model to -61.6% on the 84B model then just -3.9% on the 562B model. Felt like we were just shy of an insig…
I think our experiments actually don't show catastrophic forgetting! The accuracy does not decrease as loss gets worse -- it's simply getting over-confident.
So I'm not even sure we're showing any problem to solve here -- it might be more of a opportunity, in fact!
Re: Can LLMs learn from a single example?
#5Earlier quoted context omitted.
In the palm-e paper ( https://palm-e.github.io/ ), when they try to unfreeze and train the LLM on new image data only, there is expectedly a lot of CF on NLP tasks but very interestingly, the effect diminishes greatly with the scale of the LLM prior to training. From an average -87.3% performance drop on the 12B model to -61.6% on the 84B model then just -3.9% on the 562B model. Felt like we were just shy of an insig…
I think our experiments actually don't show catastrophic forgetting! The accuracy does not decrease as loss gets worse -- it's simply getting over-confident. So I'm not even sure we're showing any problem to solve here -- it might be more of a opportunity, in fact!
It does start getting worse at some point right?
Re: Can LLMs learn from a single example?
#6Thank you for posting this to HN! :D I'm one of the authors of this post -- Johno & I found it really interesting looking into this curious issue of rapid memorization from LLMs. I've been working with neural nets for 30 years, and fine-tuning language models since 2017, and this behavior is most surprising to me! Other folks have seen it in LLMs too, although I haven't seen a analysis of this kind before (although w…
In the palm-e paper ( https://palm-e.github.io/ ), when they try to unfreeze and train the LLM on new image data only, there is expectedly a lot of CF on NLP tasks but very interestingly, the effect diminishes greatly with the scale of the LLM prior to training. From an average -87.3% performance drop on the 12B model to -61.6% on the 84B model then just -3.9% on the 562B model. Felt like we were just shy of an insig…
Ooh interesting, thanks for sharing!
Re: Can LLMs learn from a single example?
#7Earlier quoted context omitted.
I think our experiments actually don't show catastrophic forgetting! The accuracy does not decrease as loss gets worse -- it's simply getting over-confident. So I'm not even sure we're showing any problem to solve here -- it might be more of a opportunity, in fact!
It does start getting worse at some point right?
I'm sure eventually it would, but we haven't gotten to that point yet in our training.
Re: Can LLMs learn from a single example?
#8Thank you for posting this to HN! :D I'm one of the authors of this post -- Johno & I found it really interesting looking into this curious issue of rapid memorization from LLMs. I've been working with neural nets for 30 years, and fine-tuning language models since 2017, and this behavior is most surprising to me! Other folks have seen it in LLMs too, although I haven't seen a analysis of this kind before (although w…
Super interesting! Another area that I've seen these types of loss curves are recommendation models: https://arxiv.org/pdf/2209.06053.pdf
Re: Can LLMs learn from a single example?
#9Does anyone know if LLMs have been used to augment their own training data?
I wonder what would happen if you trained an LLM on a little input but then had it generate a lot of synthetic input added to the training data. I think of it as "dreaming". This seems like it would just add noise, but LLMs are able to improve their output by augmenting their own context (by "thinking out loud"), maybe they can do the same with their own training data?
Re: Can LLMs learn from a single example?
#10If this holds true, this would support the idea that much smaller, human curated datasets will be of much higher value than synthetic datasets generated by LLMs