Singularity

218 readers

1 users here now

The technological singularity—or simply the singularity—is a hypothetical future point in time at which technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization. According to the most popular version of the singularity hypothesis, I. J. Good's intelligence explosion model, an upgradable intelligent agent will eventually enter a "runaway reaction" of self-improvement cycles, each new and more intelligent generation appearing more and more rapidly, causing an "explosion" in intelligence and resulting in a powerful superintelligence that qualitatively far surpasses all human intelligence.

— Wikipedia

This is a community for discussing theoretical and practical consequences related to the singularity, or any other innovation in the realm of machine learning capable of potentially disrupting our society.

You can share news, research papers, discussions and opinions. This community is mainly meant for information and discussion, so entertainment (such as memes) should generally be avoided, unless the content is thought-provoking or has some other qualities.

Rules:

Be nice to everyone, even if you disagree.
No spam. No ads.
No NSFW.
Self-promotion is acceptable if not excessive (i.e. no spam).

founded 1 year ago

MODERATORS

[email protected]

Reinforced Self-Training (ReST) for Language Modeling (arxiv.org)

submitted 1 year ago by [email protected] to c/[email protected]

0 comments fedilink hide all child comments

Abstract

Reinforcement learning from human feedback (RLHF) can improve the quality of large language model's (LLM) outputs by aligning them with human preferences. We propose a simple algorithm for aligning LLMs with human preferences inspired by growing batch reinforcement learning (RL), which we call Reinforced Self-Training (ReST). Given an initial LLM policy, ReST produces a dataset by generating samples from the policy, which are then used to improve the LLM policy using offline RL algorithms. ReST is more efficient than typical online RLHF methods because the training dataset is produced offline, which allows data reuse. While ReST is a general approach applicable to all generative learning settings, we focus on its application to machine translation. Our results show that ReST can substantially improve translation quality, as measured by automated metrics and human evaluation on machine translation benchmarks in a compute and sample-efficient manner.

no comments (yet)

sorted by: hot top controversial new old

there doesn't seem to be anything here