Debate Championship for LLMs
โข
4
Clelia (Astra) Bertelli PRO
as-cle-bert
AI & ML interests
Biology + Artificial Intelligence = โค๏ธ | AI for sustainable development, sustainable development for AI | Researching on Machine Learning Enhancement | I love automation for everyday things | Blogger | Open Source
Recent Activity
posted
an
update
about 12 hours ago
Are you using Obsidian to write your notes?
If the answer is yes, then this post might be for you!โ
I recently created ๐จ๐๐ฌ๐ข๐๐ข๐๐ง-๐๐ข๐ ๐๐ฌ๐ญ, a Google Gemini-powered application that gives you feedback on style and contents of the documents you have been working on๐ง
Repo ๐ https://github.com/AstraBert/obsidian-digest
PyPi Package ๐ https://pypi.org/project/obsidian-digest/
The app is available as:
- ๐๐จ๐ฆ๐ฆ๐๐ง๐-๐ฅ๐ข๐ง๐ ๐ญ๐จ๐จ๐ฅ: install it as a python package with ๐ฝ๐ถ๐ฝ, and execute it from terminal anytime!๐ฆ
-๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐ฎ๐ข๐ฅ๐ญ ๐๐ซ๐จ๐ฆ ๐ฌ๐จ๐ฎ๐ซ๐๐ ๐๐จ๐๐: clone the GitHub repo, install the needed dependencies through ๐ฐ๐ผ๐ป๐ฑ๐ฎ, and run the bot: you will get hourly messages with suggestions and considerations about your activity on Obsidian in the previous hour๐ค
- ๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐๐ฉ๐ฅ๐จ๐ฒ๐๐ ๐ฅ๐จ๐๐๐ฅ๐ฅ๐ฒ ๐ฐ๐ข๐ญ๐ก ๐๐จ๐๐ค๐๐ซ ๐๐จ๐ฆ๐ฉ๐จ๐ฌ๐: clone the GitHub repo and launch ๐ฑ๐ผ๐ฐ๐ธ๐ฒ๐ฟ ๐ฐ๐ผ๐บ๐ฝ๐ผ๐๐ฒ ๐๐ฝ. Docker builds an image on the fly with all the needed dependencies and scripts, and runs them. You'll have the same functionalities as the ones from source code, but with a way easier deployment process๐
Go check out the GitHub repo for more info ๐ https://github.com/AstraBert/obsidian-digest
Have fun!โจ
replied to
their
post
2 days ago
Hi HF Community!๐ค
As my last 2024 contribution, I decided to write an article about a Competitive Debate Championship simulation I ran with 5 LLMs as competitors and 2 as judges:
https://huggingface.co/blog/as-cle-bert/debate-championship-for-llms
The article covers code, analyses and results, and you can find everything to reproduce this tournament in the GitHub repo ๐ https://github.com/AstraBert/DebateLLM-Championship
I also released a dataset related to the data (motions, arguments, topics, winners...) collected during the tournament ๐ https://huggingface.co/datasets/as-cle-bert/DebateLLMs
Happy reading and happy new yeAIr!๐
replied to
their
post
2 days ago
๐๐๐๐ซ๐ฅ๐ฒ ๐๐๐ฐ ๐๐๐๐ซ ๐ซ๐๐ฅ๐๐๐ฌ๐๐ฌ๐
Hi HuggingFacers๐ค, I decided to ship early this year, and here's what I came up with:
๐๐๐๐๐ญ๐๐จ๐ฐ๐ง (https://github.com/AstraBert/PdfItDown) - If you're like me, and you have all your RAG pipeline optimized for PDFs, but not for other data formats, here is your solution! With PdfItDown, you can convert Word documents, presentations, HTML pages, markdown sheets and (why not?) CSVs and XMLs in PDF format, for seamless integration with your RAG pipelines. Built upon MarkItDown by Microsoft
GitHub Repo ๐ https://github.com/AstraBert/PdfItDown
PyPi Package ๐ https://pypi.org/project/pdfitdown/
๐๐๐ง๐๐ซ๐๐ฏ ๐ฏ๐.๐.๐ (https://github.com/AstraBert/SenTrEv/tree/v1.0.0) - If you need to evaluate the ๐ฟ๐ฒ๐๐ฟ๐ถ๐ฒ๐๐ฎ๐น performance of your ๐๐ฒ๐
๐ ๐ฒ๐บ๐ฏ๐ฒ๐ฑ๐ฑ๐ถ๐ป๐ด models, I have good news for you๐ฅณ๐ฅณ
The new release for ๐๐๐ง๐๐ซ๐๐ฏ now supports ๐ฑ๐ฒ๐ป๐๐ฒ and ๐๐ฝ๐ฎ๐ฟ๐๐ฒ retrieval (thanks to FastEmbed by Qdrant) with ๐๐ฒ๐
๐-๐ฏ๐ฎ๐๐ฒ๐ฑ ๐ณ๐ถ๐น๐ฒ ๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ (.docx, .pptx, .csv, .html, .xml, .md, .pdf) and new ๐ฟ๐ฒ๐น๐ฒ๐๐ฎ๐ป๐ฐ๐ฒ ๐บ๐ฒ๐๐ฟ๐ถ๐ฐ๐!
GitHub repo ๐ https://github.com/AstraBert/SenTrEv
Release Notes ๐ https://github.com/AstraBert/SenTrEv/releases/tag/v1.0.0
PyPi Package ๐ https://pypi.org/project/sentrev/
Happy New Year and have fun!๐ฅ
Articles
Organizations
as-cle-bert's activity
posted
an
update
about 12 hours ago
Post
193
Are you using Obsidian to write your notes?
If the answer is yes, then this post might be for you!โ
I recently created ๐จ๐๐ฌ๐ข๐๐ข๐๐ง-๐๐ข๐ ๐๐ฌ๐ญ, a Google Gemini-powered application that gives you feedback on style and contents of the documents you have been working on๐ง
Repo ๐ https://github.com/AstraBert/obsidian-digest
PyPi Package ๐ https://pypi.org/project/obsidian-digest/
The app is available as:
- ๐๐จ๐ฆ๐ฆ๐๐ง๐-๐ฅ๐ข๐ง๐ ๐ญ๐จ๐จ๐ฅ: install it as a python package with ๐ฝ๐ถ๐ฝ, and execute it from terminal anytime!๐ฆ
-๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐ฎ๐ข๐ฅ๐ญ ๐๐ซ๐จ๐ฆ ๐ฌ๐จ๐ฎ๐ซ๐๐ ๐๐จ๐๐: clone the GitHub repo, install the needed dependencies through ๐ฐ๐ผ๐ป๐ฑ๐ฎ, and run the bot: you will get hourly messages with suggestions and considerations about your activity on Obsidian in the previous hour๐ค
- ๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐๐ฉ๐ฅ๐จ๐ฒ๐๐ ๐ฅ๐จ๐๐๐ฅ๐ฅ๐ฒ ๐ฐ๐ข๐ญ๐ก ๐๐จ๐๐ค๐๐ซ ๐๐จ๐ฆ๐ฉ๐จ๐ฌ๐: clone the GitHub repo and launch ๐ฑ๐ผ๐ฐ๐ธ๐ฒ๐ฟ ๐ฐ๐ผ๐บ๐ฝ๐ผ๐๐ฒ ๐๐ฝ. Docker builds an image on the fly with all the needed dependencies and scripts, and runs them. You'll have the same functionalities as the ones from source code, but with a way easier deployment process๐
Go check out the GitHub repo for more info ๐ https://github.com/AstraBert/obsidian-digest
Have fun!โจ
If the answer is yes, then this post might be for you!โ
I recently created ๐จ๐๐ฌ๐ข๐๐ข๐๐ง-๐๐ข๐ ๐๐ฌ๐ญ, a Google Gemini-powered application that gives you feedback on style and contents of the documents you have been working on๐ง
Repo ๐ https://github.com/AstraBert/obsidian-digest
PyPi Package ๐ https://pypi.org/project/obsidian-digest/
The app is available as:
- ๐๐จ๐ฆ๐ฆ๐๐ง๐-๐ฅ๐ข๐ง๐ ๐ญ๐จ๐จ๐ฅ: install it as a python package with ๐ฝ๐ถ๐ฝ, and execute it from terminal anytime!๐ฆ
-๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐ฎ๐ข๐ฅ๐ญ ๐๐ซ๐จ๐ฆ ๐ฌ๐จ๐ฎ๐ซ๐๐ ๐๐จ๐๐: clone the GitHub repo, install the needed dependencies through ๐ฐ๐ผ๐ป๐ฑ๐ฎ, and run the bot: you will get hourly messages with suggestions and considerations about your activity on Obsidian in the previous hour๐ค
- ๐๐ข๐ฌ๐๐จ๐ซ๐ ๐๐จ๐ญ ๐๐๐ฉ๐ฅ๐จ๐ฒ๐๐ ๐ฅ๐จ๐๐๐ฅ๐ฅ๐ฒ ๐ฐ๐ข๐ญ๐ก ๐๐จ๐๐ค๐๐ซ ๐๐จ๐ฆ๐ฉ๐จ๐ฌ๐: clone the GitHub repo and launch ๐ฑ๐ผ๐ฐ๐ธ๐ฒ๐ฟ ๐ฐ๐ผ๐บ๐ฝ๐ผ๐๐ฒ ๐๐ฝ. Docker builds an image on the fly with all the needed dependencies and scripts, and runs them. You'll have the same functionalities as the ones from source code, but with a way easier deployment process๐
Go check out the GitHub repo for more info ๐ https://github.com/AstraBert/obsidian-digest
Have fun!โจ
replied to
their
post
2 days ago
Hi and thanks a lot for the specification!๐ฅฐ
Just as a note from my side, in the article I specify that there is a difference between "open weights" and "open source" models, and I link this blog post: https://www.agora.software/en/llm-open-source-open-weight-or-proprietary/ for a deeper explanation of the difference. I never (and I would never) claimed that Llama is open source, let alone a free software (see the introduction in this article of mine on privacy and data "stealing" risks: https://huggingface.co/blog/as-cle-bert/build-an-ai-powered-search-engine-from-scratch).
And I would have gladly used also DeepSeek, if it had been available on HuggingChat! :)
I nevertheless highly appreciate your comment and I'll for sure be more cautious in using the word "open/open source" in the future. Thanks!โจ
replied to
their
post
2 days ago
Both PdfItDown and SenTrEv only work with text for now: in future releases, support for image will be added :)
For text extraction, I use PyPDF + Langchain
posted
an
update
3 days ago
Post
1918
๐๐๐๐ซ๐ฅ๐ฒ ๐๐๐ฐ ๐๐๐๐ซ ๐ซ๐๐ฅ๐๐๐ฌ๐๐ฌ๐
Hi HuggingFacers๐ค, I decided to ship early this year, and here's what I came up with:
๐๐๐๐๐ญ๐๐จ๐ฐ๐ง (https://github.com/AstraBert/PdfItDown) - If you're like me, and you have all your RAG pipeline optimized for PDFs, but not for other data formats, here is your solution! With PdfItDown, you can convert Word documents, presentations, HTML pages, markdown sheets and (why not?) CSVs and XMLs in PDF format, for seamless integration with your RAG pipelines. Built upon MarkItDown by Microsoft
GitHub Repo ๐ https://github.com/AstraBert/PdfItDown
PyPi Package ๐ https://pypi.org/project/pdfitdown/
๐๐๐ง๐๐ซ๐๐ฏ ๐ฏ๐.๐.๐ (https://github.com/AstraBert/SenTrEv/tree/v1.0.0) - If you need to evaluate the ๐ฟ๐ฒ๐๐ฟ๐ถ๐ฒ๐๐ฎ๐น performance of your ๐๐ฒ๐ ๐ ๐ฒ๐บ๐ฏ๐ฒ๐ฑ๐ฑ๐ถ๐ป๐ด models, I have good news for you๐ฅณ๐ฅณ
The new release for ๐๐๐ง๐๐ซ๐๐ฏ now supports ๐ฑ๐ฒ๐ป๐๐ฒ and ๐๐ฝ๐ฎ๐ฟ๐๐ฒ retrieval (thanks to FastEmbed by Qdrant) with ๐๐ฒ๐ ๐-๐ฏ๐ฎ๐๐ฒ๐ฑ ๐ณ๐ถ๐น๐ฒ ๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ (.docx, .pptx, .csv, .html, .xml, .md, .pdf) and new ๐ฟ๐ฒ๐น๐ฒ๐๐ฎ๐ป๐ฐ๐ฒ ๐บ๐ฒ๐๐ฟ๐ถ๐ฐ๐!
GitHub repo ๐ https://github.com/AstraBert/SenTrEv
Release Notes ๐ https://github.com/AstraBert/SenTrEv/releases/tag/v1.0.0
PyPi Package ๐ https://pypi.org/project/sentrev/
Happy New Year and have fun!๐ฅ
Hi HuggingFacers๐ค, I decided to ship early this year, and here's what I came up with:
๐๐๐๐๐ญ๐๐จ๐ฐ๐ง (https://github.com/AstraBert/PdfItDown) - If you're like me, and you have all your RAG pipeline optimized for PDFs, but not for other data formats, here is your solution! With PdfItDown, you can convert Word documents, presentations, HTML pages, markdown sheets and (why not?) CSVs and XMLs in PDF format, for seamless integration with your RAG pipelines. Built upon MarkItDown by Microsoft
GitHub Repo ๐ https://github.com/AstraBert/PdfItDown
PyPi Package ๐ https://pypi.org/project/pdfitdown/
๐๐๐ง๐๐ซ๐๐ฏ ๐ฏ๐.๐.๐ (https://github.com/AstraBert/SenTrEv/tree/v1.0.0) - If you need to evaluate the ๐ฟ๐ฒ๐๐ฟ๐ถ๐ฒ๐๐ฎ๐น performance of your ๐๐ฒ๐ ๐ ๐ฒ๐บ๐ฏ๐ฒ๐ฑ๐ฑ๐ถ๐ป๐ด models, I have good news for you๐ฅณ๐ฅณ
The new release for ๐๐๐ง๐๐ซ๐๐ฏ now supports ๐ฑ๐ฒ๐ป๐๐ฒ and ๐๐ฝ๐ฎ๐ฟ๐๐ฒ retrieval (thanks to FastEmbed by Qdrant) with ๐๐ฒ๐ ๐-๐ฏ๐ฎ๐๐ฒ๐ฑ ๐ณ๐ถ๐น๐ฒ ๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ (.docx, .pptx, .csv, .html, .xml, .md, .pdf) and new ๐ฟ๐ฒ๐น๐ฒ๐๐ฎ๐ป๐ฐ๐ฒ ๐บ๐ฒ๐๐ฟ๐ถ๐ฐ๐!
GitHub repo ๐ https://github.com/AstraBert/SenTrEv
Release Notes ๐ https://github.com/AstraBert/SenTrEv/releases/tag/v1.0.0
PyPi Package ๐ https://pypi.org/project/sentrev/
Happy New Year and have fun!๐ฅ
-
- 2 replies
reacted to
nroggendorff's
post with โ
3 days ago
Post
5078
hey nvidia, can you send me a gpu?
comment or react if you want ~~me~~ to get one too. ๐๐
comment or react if you want ~~me~~ to get one too. ๐๐
- 22 replies
posted
an
update
5 days ago
Post
509
Hi HF Community!๐ค
As my last 2024 contribution, I decided to write an article about a Competitive Debate Championship simulation I ran with 5 LLMs as competitors and 2 as judges:
https://huggingface.co/blog/as-cle-bert/debate-championship-for-llms
The article covers code, analyses and results, and you can find everything to reproduce this tournament in the GitHub repo ๐ https://github.com/AstraBert/DebateLLM-Championship
I also released a dataset related to the data (motions, arguments, topics, winners...) collected during the tournament ๐ as-cle-bert/DebateLLMs
Happy reading and happy new yeAIr!๐
As my last 2024 contribution, I decided to write an article about a Competitive Debate Championship simulation I ran with 5 LLMs as competitors and 2 as judges:
https://huggingface.co/blog/as-cle-bert/debate-championship-for-llms
The article covers code, analyses and results, and you can find everything to reproduce this tournament in the GitHub repo ๐ https://github.com/AstraBert/DebateLLM-Championship
I also released a dataset related to the data (motions, arguments, topics, winners...) collected during the tournament ๐ as-cle-bert/DebateLLMs
Happy reading and happy new yeAIr!๐
-
-
- 3 replies
upvoted
an
article
5 days ago
Article
Debate Championship for LLMs
By
โข
โข
4
published
an
article
5 days ago
Article
Debate Championship for LLMs
By
โข
โข
4
updated
a
dataset
5 days ago
liked
a
dataset
6 days ago
liked
2
models
6 days ago
posted
an
update
9 days ago
Post
2149
I got my GitHub Wrapped for 2024 today!๐ฅ
Get yours here on HuggingFace ๐ as-cle-bert/what-a-git-year
GitHub repo with the code to reproduce it ๐ https://github.com/AstraBert/what-a-git-year
Hope that everybody had a Git year!๐
Get yours here on HuggingFace ๐ as-cle-bert/what-a-git-year
GitHub repo with the code to reproduce it ๐ https://github.com/AstraBert/what-a-git-year
Hope that everybody had a Git year!๐
- 1 reply
updated
2
Spaces
10 days ago
posted
an
update
11 days ago
Post
1700
Hi HuggingFacers!๐คถ๐ผ
As my last 2024 project, I've dropped a Discord Bot that knows a lot about Pokemons๐ฆ
GitHub ๐ https://github.com/AstraBert/Pokemon-Bot
Demo Space ๐ as-cle-bert/pokemon-bot
The bot integrates:
- Chat features (Cohere's Command-R) with RAG functionalities (hybrid search and reranking with Qdrant) and chat memory (managed through PostgreSQL) to produce information about Pokemons
- Image-based search to identify Pokemons from their images (via Qdrant)
- Card package random extraction and description
HuggingFace๐ค, as usual, plays the most important role in the application stack, with the following models:
- sentence-transformers/LaBSE
- prithivida/Splade_PP_en_v1
- facebook/dinov2-large
And datasets:
- Karbo31881/Pokemon_images
- wanghaofan/pokemon-wiki-captions
- TheFusion21/PokemonCards
Have fun!๐
As my last 2024 project, I've dropped a Discord Bot that knows a lot about Pokemons๐ฆ
GitHub ๐ https://github.com/AstraBert/Pokemon-Bot
Demo Space ๐ as-cle-bert/pokemon-bot
The bot integrates:
- Chat features (Cohere's Command-R) with RAG functionalities (hybrid search and reranking with Qdrant) and chat memory (managed through PostgreSQL) to produce information about Pokemons
- Image-based search to identify Pokemons from their images (via Qdrant)
- Card package random extraction and description
HuggingFace๐ค, as usual, plays the most important role in the application stack, with the following models:
- sentence-transformers/LaBSE
- prithivida/Splade_PP_en_v1
- facebook/dinov2-large
And datasets:
- Karbo31881/Pokemon_images
- wanghaofan/pokemon-wiki-captions
- TheFusion21/PokemonCards
Have fun!๐
liked
a
Space
11 days ago
liked
a
dataset
12 days ago
liked
a
dataset
16 days ago
posted
an
update
24 days ago
Post
598
Hi HF Community!
I just published a blog article on building PrAIvateSearch (https://github.com/AstraBert/PrAIvateSearch), a user-owend, local and open-source AI-powered search engine๐:
https://huggingface.co/blog/as-cle-bert/build-an-ai-powered-search-engine-from-scratch
"Own your AI, search the web with it๐๐"
Feel free to try it out and contribute to it on GitHub: let's make OSS AI grown and thrive!๐
I just published a blog article on building PrAIvateSearch (https://github.com/AstraBert/PrAIvateSearch), a user-owend, local and open-source AI-powered search engine๐:
https://huggingface.co/blog/as-cle-bert/build-an-ai-powered-search-engine-from-scratch
"Own your AI, search the web with it๐๐"
Feel free to try it out and contribute to it on GitHub: let's make OSS AI grown and thrive!๐