Semanlink - [2302.04761] Toolformer: Language Models Can Teach Themselves to Use Tools

Tags:

> Toolformer, **a model
trained to decide which APIs to call, when to
call them, what arguments to pass, and how to
best incorporate the results into future token
prediction**.

> fulfills the
following desiderata:
> - The use of tools should be learned in a
self-supervised way without requiring large
amounts of human annotations
>- The LM should be able to decide for itself when
and how to use which tool.

> Approach based
on the recent idea of using large LMs with incontext
learning (Brown et al., 2020) to generate
entire datasets from scratch.
>
> Given just a handful of human-written examples
of how an API can be used, 
> - we let a LM annotate
a huge language modeling dataset with potential
API calls. 
> - We then use a self-supervised loss to
determine which of these API calls actually help
the model in predicting future tokens. 
>- Finally, we
finetune the LM itself on the API calls that it considers
useful.

[Jay Hack @mathemagic1an sur twitter](https://twitter.com/mathemagic1an/status/1624870248221663232):

> from a small seed set of human inputs (essentially demonstrating usage of APIs), the training set for this behavior is generated by the LLM itself.
> 
> So what does this mean? We've found a promising way to tightly integrate arbitrary APIs with our best-performing models.

Au sujet de ce document

sl:arxiv_author :
sl:arxiv_firstAuthor : Timo Schick
sl:arxiv_num : 2302.04761
sl:arxiv_published : 2023-02-09T16:49:57Z
sl:arxiv_summary : Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale. They also, paradoxically, struggle with basic functionality, such as arithmetic or factual lookup, where much simpler and smaller models excel. In this paper, we show that LMs can teach themselves to use external tools via simple APIs and achieve the best of both worlds. We introduce Toolformer, a model trained to decide which APIs to call, when to call them, what arguments to pass, and how to best incorporate the results into future token prediction. This is done in a self-supervised way, requiring nothing more than a handful of demonstrations for each API. We incorporate a range of tools, including a calculator, a Q\&A system, two different search engines, a translation system, and a calendar. Toolformer achieves substantially improved zero-shot performance across a variety of downstream tasks, often competitive with much larger models, without sacrificing its core language modeling abilities.@en
sl:arxiv_title : Toolformer: Language Models Can Teach Themselves to Use Tools@en
sl:arxiv_updated : 2023-02-09T16:49:57Z
sl:bookmarkOf : https://arxiv.org/abs/2302.04761
sl:creationDate : 2023-02-13
sl:creationTime : 2023-02-13T15:18:25Z
sl:relatedDoc : http://www.semanlink.net/doc/2023/02/jay_hack_sur_twitter_my_thou

Infos sur le fichier

Bookmark of: https://arxiv.org/abs/2302.04761

Documents with similar tags (experimental)

[2310.01352] RA-DIT: Retrieval-Augmented Dual Instruction Tuning

Tags:

2023-10-07 A propos

[2308.13418] Nougat: Neural Optical Understanding for Academic Documents

Tags:

2023-09-17 A propos

[2305.07185] MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers

Tags:

2023-07-01 A propos

[2303.14177] Scaling Expert Language Models with Unsupervised Domain Discovery

Tags:

2023-03-27 A propos

[2212.09741] One Embedder, Any Task: Instruction-Finetuned Text Embeddings

Tags:

2023-02-17 A propos