Large Language Models Can Self-Improve · HackerPapers discussions

Large Language Models Can Self-Improve — mentioned by taneq

Tue, 09 May 2023 13:35:17 GMT

> This result feels like creepy perpetual motion. It’s like they’re teaching the AI ethics by making it write an ethics textbook and then read the textbook it just wrote. Is this a free lunch? Shouldn’t it be impossible for the AI to teach itself any more ethics than it started out with? Isn't this kind of feedback loop precisely what's predicted at the heart of the Singularity? A machine learning system reaching the point where it's able to improve itself without further human guidance? "Large Language Models Can Self-Improve" (https://arxiv.org/abs/2210.11610) describes something similar…

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by kzrdude

Tue, 07 Mar 2023 10:00:10 GMT

I don't think it can be so easily dismissed. AlphaZero contributed lots, for example reproducing opening theory from a blank slate. For chess that's significant. But we don't care about chess for this discussion about GPT. Self-learning and self-play would be interesting because it's an amplification factor. Some searching suggests some researchers have been thinking along these lines. Maybe this? https://arxiv.org/abs/2210.11610 Going for methods that don't require supervision

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by jbenjoseph

Fri, 24 Feb 2023 04:52:04 GMT

I have seen no evidence for that, only the opposite: https://arxiv.org/abs/2210.11610 Intuitively, training a simple enough linear statistical model with its own output should be a NOP. But LLMs are anything but simple models, so I think the non-linearities may be synthesizing new useful information. Similarly to how all of maths can be synthesized from a few basic axioms with enough intelligence or computation.

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by sbierwagen

Wed, 22 Feb 2023 04:29:47 GMT

Here's a LLM fine-tuned on its own output that improved on several tasks: https://arxiv.org/abs/2210.11610

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by moyix

Wed, 22 Feb 2023 00:59:23 GMT

Here’s a more directly comparable example: https://arxiv.org/abs/2210.11610

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by valine

Mon, 13 Feb 2023 14:52:56 GMT

There is evidence to suggest that LLMs have recently passed a threshold where their performance can be improved by training on their own output. https://arxiv.org/abs/2210.11610 I expect your concerns are likely unwarranted.

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by anabis

Mon, 13 Feb 2023 14:28:59 GMT

>But, it's not implausible that some future model will simply have a superhuman learning rate and a superhuman ability to distinguish "right" from "wrong" - this paragraph will look downright prophetic then. There is already a paper for that: https://arxiv.org/abs/2210.11610 Large Language Models Can Self-Improve >Large Language Models (LLMs) have achieved excellent performances in various tasks. However, fine-tuning an LLM requires extensive supervision. Human, on the other hand, may improve their reasoning abilities by self-thinking without external inputs. In this work, we demonstrate that…

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — HN submission

Sun, 12 Feb 2023 13:17:52 GMT

Large Language Models Can Self-Improve

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Large Language Models Can Self-Improve — HN submission

Sun, 05 Feb 2023 13:57:32 GMT

Large language models can self-improve

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Large Language Models Can Self-Improve — mentioned by vacuumcl

Fri, 03 Feb 2023 17:39:51 GMT

There are ways that LLM's can self-improve, such as in this paper: https://arxiv.org/abs/2210.11610 I would speculate that there are more ways to train on logical consistency of the output, and improve the models further.

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — mentioned by visarga

Fri, 13 Jan 2023 17:07:32 GMT

It was not a language model, but AlphaZero played against itself millions of games and trained on the data it generated, without seeing any human games. And it reached super human level, of course. Why was this possible? The game itself acted as an anti-bullshit filter. So you can train on your own generated data if you filter it. Like this one: Large Language Models Can Self Improve https://arxiv.org/abs/2210.11610

Paper and discussion history. A mention is not an endorsement.

Large Language Models Can Self-Improve — HN submission

Tue, 01 Nov 2022 20:22:17 GMT

Large Language Models Can Self-Improve

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.