Daniel Leong

daniel-ltw

daniel-ltw

AI & ML interests

None yet

Recent Activity

reacted to burtenshaw's post with 🚀 7 days ago

AGENTS + FINETUNING! This week Hugging Face learn has a whole pathway on finetuning for agentic applications. You can follow these two courses to get knowledge on levelling up your agent game beyond prompts: 1️⃣ New Supervised Fine-tuning unit in the NLP Course https://huggingface.co/learn/nlp-course/en/chapter11/1 2️⃣New Finetuning for agents bonus module in the Agents Course https://huggingface.co/learn/agents-course/bonus-unit1/introduction Fine-tuning will squeeze everything out of your model for how you’re using it, more than any prompt.

reacted to fdaudens's post with ❤️ 8 days ago

🎯 Perplexity drops their FIRST open-weight model on Hugging Face: A decensored DeepSeek-R1 with full reasoning capabilities. Tested on 1000+ examples for unbiased responses. Check it out: https://huggingface.co/perplexity-ai/r1-1776 Blog post: https://perplexity.ai/hub/blog/open-sourcing-r1-1776

reacted to fdaudens's post with 👍 8 days ago

View all activity

Organizations

None yet

daniel-ltw's activity

reacted to burtenshaw's post with 🚀 7 days ago

Post

6815

AGENTS + FINETUNING! This week Hugging Face learn has a whole pathway on finetuning for agentic applications. You can follow these two courses to get knowledge on levelling up your agent game beyond prompts:

1️⃣ New Supervised Fine-tuning unit in the NLP Course https://huggingface.co/learn/nlp-course/en/chapter11/1
2️⃣New Finetuning for agents bonus module in the Agents Course https://huggingface.co/learn/agents-course/bonus-unit1/introduction

Fine-tuning will squeeze everything out of your model for how you’re using it, more than any prompt.

2 replies

reacted to fdaudens's post with ❤️👍 8 days ago

Post

5746

🎯 Perplexity drops their FIRST open-weight model on Hugging Face: A decensored DeepSeek-R1 with full reasoning capabilities. Tested on 1000+ examples for unbiased responses.

Check it out: perplexity-ai/r1-1776
Blog post: https://perplexity.ai/hub/blog/open-sourcing-r1-1776

1 reply

reacted to Xenova's post with 🔥 17 days ago

Post

7532

We did it. Kokoro TTS (v1.0) can now run 100% locally in your browser w/ WebGPU acceleration. Real-time text-to-speech without a server. ⚡️

Generate 10 seconds of speech in ~1 second for $0.

What will you build? 🔥
webml-community/kokoro-webgpu

The most difficult part was getting the model running in the first place, but the next steps are simple:
✂️ Implement sentence splitting, allowing for streamed responses
🌍 Multilingual support (only phonemization left)

Who wants to help?

9 replies

reacted to Kseniase's post with 🔥 17 days ago

Post

7688

8 New Types of RAG

RAG techniques continuously evolve to enhance LLM response accuracy by retrieving relevant external data during generation. To keep up with current AI trends, new RAG types incorporate deep step-by-step reasoning, tree search, citations, multimodality and other effective techniques.

Here's a list of 8 latest RAG advancements:

1. DeepRAG -> DeepRAG: Thinking to Retrieval Step by Step for Large Language Models (2502.01142)
Models retrieval-augmented reasoning as a Markov Decision Process, enabling strategic retrieval. It dynamically decides when to retrieve external knowledge and when rely on parametric reasoning.

2. RealRAG -> RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive Learning (2502.00848)
Enhances novel object generation by retrieving real-world images and using self-reflective contrastive learning to fill knowledge gap, improve realism and reduce distortions.

3. Chain-of-Retrieval Augmented Generation (CoRAG) -> Chain-of-Retrieval Augmented Generation (2501.14342)
Retrieves information step-by-step and adjusts it, also deciding how much compute power to use at test time. If needed it reformulates queries.

4. VideoRAG -> VideoRAG: Retrieval-Augmented Generation over Video Corpus (2501.05874)
Enables unlimited-length video processing, using dual-channel architecture that integrates graph-based textual grounding and multi-modal context encoding.

5. CFT-RAG -> CFT-RAG: An Entity Tree Based Retrieval Augmented Generation Algorithm With Cuckoo Filter (2501.15098)
A tree-RAG acceleration method uses an improved Cuckoo Filter to optimize entity localization, enabling faster retrieval.

6. Contextualized Graph RAG (CG-RAG) -> CG-RAG: Research Question Answering by Citation Graph Retrieval-Augmented LLMs (2501.15067)
Uses Lexical-Semantic Graph Retrieval (LeSeGR) to integrate sparse and dense signals within graph structure and capture citation relationships

7. GFM-RAG -> GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (2502.01113)
A graph foundation model that uses a graph neural network to refine query-knowledge connections

8. URAG -> URAG: Implementing a Unified Hybrid RAG for Precise Answers in University Admission Chatbots -- A Case Study at HCMUT (2501.16276)
A hybrid system combining rule-based and RAG methods to improve lightweight LLMs for educational chatbots

1 reply

reacted to singhsidhukuldeep's post with 🔥 23 days ago

Post

2413

Excited to share groundbreaking research in Knowledge Graph-based Retrieval-Augmented Generation (KG-RAG)!

Researchers from the University of Science and Technology of China have developed FRAG - a novel flexible modular framework that revolutionizes how Large Language Models (LLMs) reason with knowledge graphs.

What makes FRAG special? It intelligently adapts retrieval strategies based on query complexity without requiring expensive KG fine-tuning. The framework uses a reasoning-aware module to classify queries as simple or complex, then applies tailored retrieval pipelines.

Under the hood:
- For simple queries: Uses breadth-first search and ranking to efficiently find relevant paths
- For complex queries: Employs shortest path algorithms to minimize computational overhead
- Features a preprocessing-retrieval-postprocessing pipeline with flexible components
- Leverages traditional algorithms like PersonalizedPageRank for subgraph extraction
- Implements edge and path ranking models for precise information filtering

The results are impressive - FRAG achieves state-of-the-art performance while maintaining high efficiency and low resource consumption. On benchmark datasets like WebQSP and CWQ, it outperforms existing approaches by significant margins.

Most importantly, FRAG maintains flexibility and modularity while improving retrieval quality - no expensive LLM fine-tuning required! This makes it highly practical for real-world applications.

This work represents a major step forward in making LLMs more reliable and capable of complex reasoning tasks. Looking forward to seeing how this technology evolves!

2 replies

reacted to m-ric's post with 🔥 24 days ago

Post

3102

Now you can launch a code agent directly from your terminal!
✨ 𝚜𝚖𝚘𝚕𝚊𝚐𝚎𝚗𝚝 "𝚈𝚘𝚞𝚛 𝚝𝚊𝚜𝚔" directly launches a CodeAgent
▶️ This also works with web agents (replace 𝚜𝚖𝚘𝚕𝚊𝚐𝚎𝚗𝚝 with 𝚠𝚎𝚋𝚊𝚐𝚎𝚗𝚝) thanks to @merve !

💾 Another treat from smolagents release 1.7.0:
Now agents have a memory mechanism, enabling many possibilities like replaying the last run with 𝚊𝚐𝚎𝚗𝚝.𝚛𝚎𝚙𝚕𝚊𝚢(), thank you @clefourrier !

Check the release notes here 👉 https://github.com/huggingface/smolagents/releases/tag/v1.7.0

replied to etemiz's post about 1 month ago

I guess when you say beneficial to humans, that could also be subjective?

Like we can now say this vitamin or medication has benefits that over weigh the negatives, but this could also be due to full studies might not have been done to prove the other negatives that could come with it. We are just weighing heavily on what we know base on what we have seen or heard.

Also, with the above, as the human genome is different, benefits to one might not be the same to others.

I do reckon, in such situation, an AI not taking a side to be a better approach and prompt the humans to do their own research. I'm also pretty sure this medication example do also applies to other paradigms/areas.

replied to etemiz's post about 1 month ago

Define human alignment. Is human alignment, what the majority says? Is majority always correct?

These are critical questions that need to also be considered.

reacted to sagar007's post with 🔥 about 1 month ago

Post

3541

🚀 Just built a Perplexity-inspired AI search assistant using Gradio, DeepSeek, and DuckDuckGo!
Ask it anything, and it’ll:

Scour the web for answers 📚

Cite sources like a pro 🔗

Even talk back with TTS (thanks, Kokoro!) 🎙️

Ask it anything, and it’ll:

Scour the web for answers 📚

Check it out → sagar007/DeepSeekR1_Search

reacted to singhsidhukuldeep's post with 👍 about 1 month ago

Post

598

Exciting breakthrough in Text Embeddings: Introducing LENS (Lexicon-based EmbeddiNgS)!

A team of researchers from University of Amsterdam, University of Technology Sydney, and Tencent have developed a groundbreaking approach that outperforms dense embeddings on the Massive Text Embedding Benchmark (MTEB).

>> Key Technical Innovations:
- LENS consolidates vocabulary space through token embedding clustering, addressing the inherent redundancy in LLM tokenizers
- Implements bidirectional attention and innovative pooling strategies to unlock the full potential of LLMs
- Each dimension corresponds to token clusters instead of individual tokens, creating more coherent and compact embeddings
- Achieves competitive performance with just 4,000-8,000 dimensional embeddings, matching the size of dense counterparts

>> Under the Hood:
The framework applies KMeans clustering to token embeddings from the language modeling head, replacing original embeddings with cluster centroids. This reduces dimensionality while preserving semantic relationships.

>> Results:
- Outperforms dense embeddings on MTEB benchmark
- Achieves state-of-the-art performance when combined with dense embeddings on BEIR retrieval tasks
- Demonstrates superior performance across clustering, classification, and retrieval tasks

This work opens new possibilities for more efficient and interpretable text embeddings. The code will be available soon.

1 reply

reacted to Jaward's post with 👍 about 1 month ago

Post

1875

minimal single script implementation of knowledge distillation in LLMs. In this implementation, we use GPT-2 (124M) as student model and GPT-2 Medium (340M) as teacher via reverse Kullback-Leibler (KL) divergence, trained on a small chunk of openwebtext.

Code: https://github.com/Jaykef/ai-algorithms/blob/main/llm_knowledge_distillation.ipynb

liked a Space 2 months ago

531

Open Source Ai Year In Review 2024

😻

What happened in open-source AI this year, and what’s next?

reacted to julien-c's post with 🔥 3 months ago

Post

10006

After some heated discussion 🔥, we clarify our intent re. storage limits on the Hub

TL;DR:
- public storage is free, and (unless blatant abuse) unlimited. We do ask that you consider upgrading to PRO and/or Enterprise Hub if possible
- private storage is paid above a significant free tier (1TB if you have a paid account, 100GB otherwise)

docs: https://huggingface.co/docs/hub/storage-limits

We optimize our infrastructure continuously to scale our storage for the coming years of growth in Machine learning, to the benefit of the community 🔥

cc: @reach-vb @pierric @victor and the HF team

28 replies

reacted to m-ric's post with 🔥 3 months ago

Post

2261

Last week was crazy in OS AI, with important models and datasets releases every day.

Here are the most important ones I've pinned:

🌎 Cohere relased GLobal-MMLU, a multilingual version of MMLU, to evaluate AI models' world knowledge in many languages!

🦙 Meta released Llama-3.3-70B-Instruct, a 70B model that's on par with Llama-3.1-405B-Instruct, GPT-4o and Claude. Probably my new go-to for agentic workflows.

🔉 FishAudio released fish-speech-1.5, multilingual text to speech model

🎨 Microsoft Research released TRELLIS, an extremely impressive image-to-3D model, which you can try here: JeffreyXiang/TRELLIS

📚 Yesterday, Hugging Face release FineWeb 2, a new version that extends the previous FineWeb to over 1000 languages, including extended coverage in Russina, Mandarin, German, Japanese, Spanish, French, so a huge, high-quality dataset of > 3 trillion words! HuggingFaceFW/fineweb-2

Now let's go build to make this week as productive as last one!

reacted to AdinaY's post with 🔥 3 months ago

Post

1653

🌊 The wave of reasoning models from the Chinese community has arrived!

🚀 Marco-o1 by AIDC, Alibaba
👉 AIDC-AI/Marco-o1

✨ QwQ by Qwen, Alibaba
👉 Qwen/qwq-674762b79b75eac01735070a

🌟 Skywork-o1 by Kunlun Tech
👉 Skywork/skywork-o1-open-67453df58e12f6c3934738d0

🔥 Xkev/Llama-3.2V-11B-cot by PKU Yuan group
👉 Xkev/Llama-3.2V-11B-cot

💡 DeepSeek-R1-Lite-Preview by DeepSeek AI
👉 https://chat.deepseek.com/

🔍 InternThinker Preview by Shanghai AI Lab
👉 https://sso.openxlab.org.cn/login?redirect=https://internlm-chat.intern-ai.org.cn/&clientId=ebmrvod6yo0nlzaek1yp

📘 k0-math by Moonshot AI
🚀 https://kimi.moonshot.cn/ ( coming soon! )

Who's next? 👀
zh-ai-community/reasoning-models-67409fb3aa1ed78f10087cd7

reacted to luigi12345's post with 👍 3 months ago

Post

3741

MinimalScrap
Only Free Dependencies. Save it.It is quite useful uh.

!pip install googlesearch-python requests
from googlesearch import search
import requests
query = "Glaucoma"
for url in search(f"{query} site:nih.gov filetype:pdf", 20):
    if url.endswith(".pdf"):
        with open(url.split("/")[-1], "wb") as f: f.write(requests.get(url).content)
        print("✅" + url.split("/")[-1])
print("Done!")

reacted to jjokah's post with 👍 3 months ago

Post

792

Google's revamped Machine Learning Crash Course covers the recent advances in AI, with an increased focus on interactive learning.

📝 100+ exercises
🗂 12 modules
🕒 15 hours
📹 Video explainers of ML concepts
🌎 Real-world examples
📊 Interactive visualizations

Ref:
https://developers.google.com/machine-learning/crash-course

reacted to hbseong's post with 🔥 4 months ago

Post

3304

🚨🔥 New Release Alert! 🔥🚨

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! 💻💥
👉 Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! 💡)

More details in our paper: https://arxiv.org/abs/2410.01524 📜

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning

reacted to as-cle-bert's post with 🚀 4 months ago

Post

3217

Hi HuggingFacers!🤗

As you may have probably heard, in the past weeks three Tech Giants (Microsoft, Amazon and Google) announced that they would bet on nuclear reactors to feed the surging energy demand of data centers, driven by increasing AI data and computational flows.

I try to explain the state of AI energy consumptions, its environmental impact and the key points of "turning AI nuclear" in my last article on HF community blog: https://huggingface.co/blog/as-cle-bert/ai-is-turning-nuclear-a-review

Enjoy the reading!🌱