An Open Realtime Voice You Can Actually Run Yourself
📝
22
Generate real‑time spoken AI conversations
🤝 Open to Collab
Mohammed Hamdy
mmhamdy
AI & ML interests
AI4Sci | NLP | Reinforcement Learning
Recent Activity
liked a Space 1 day ago
HuggingFaceM4/hugging-voice repliedto their post 12 days ago
Decades before the modern scaling laws, this paper showed that neural networks behavior under scale follows remarkably predictable laws.
In 1993, researchers at Bell Labs were grappling with a constraint that feels entirely familiar (and contemporary): datasets were outgrowing the available hardware, and training a model to the end was becoming too expensive. To evaluate an architectural tweak to a state-of-the-art model (at the time it was LeNet) on 60,000 samples meant burning up to three weeks of compute time.
To save compute, people would train candidate architectures on small subsets of the data, assuming that the top performer at small scale would remain the top performer at full scale. But with our future wisdom, we know this is not the case.
In "Learning Curves: Asymptotic Values and Rate of Convergence (NeurIPS 93)", using insights from statistical mechanics, they proposed a practical and principled method for predicting the performance of classifiers trained on large datasets (at the time, models were assumed to be large enough). The method was based on a simple power-law modeling of the expected training and test errors.
It is often noted that many of today's breakthroughs in AI and deep learning are actually decades-old concepts that simply lacked the computational power to be tested at the time. While there is some truth to that, it highlights a more valuable lesson: there is immense worth in revisiting early literature and reflecting on foundational ideas we may have prematurely left behind.
So, go explore and find your own inspiration. The current trend has enough champions already! posted an update 12 days ago
Decades before the modern scaling laws, this paper showed that neural networks behavior under scale follows remarkably predictable laws.
In 1993, researchers at Bell Labs were grappling with a constraint that feels entirely familiar (and contemporary): datasets were outgrowing the available hardware, and training a model to the end was becoming too expensive. To evaluate an architectural tweak to a state-of-the-art model (at the time it was LeNet) on 60,000 samples meant burning up to three weeks of compute time.
To save compute, people would train candidate architectures on small subsets of the data, assuming that the top performer at small scale would remain the top performer at full scale. But with our future wisdom, we know this is not the case.
In "Learning Curves: Asymptotic Values and Rate of Convergence (NeurIPS 93)", using insights from statistical mechanics, they proposed a practical and principled method for predicting the performance of classifiers trained on large datasets (at the time, models were assumed to be large enough). The method was based on a simple power-law modeling of the expected training and test errors.
It is often noted that many of today's breakthroughs in AI and deep learning are actually decades-old concepts that simply lacked the computational power to be tested at the time. While there is some truth to that, it highlights a more valuable lesson: there is immense worth in revisiting early literature and reflecting on foundational ideas we may have prematurely left behind.
So, go explore and find your own inspiration. The current trend has enough champions already!Organizations
liked a
Space 1 day ago
replied to their post 12 days ago
Learning Curves: Asymptotic Values and Rate of Convergence,
https://proceedings.neurips.cc/paper/1993/hash/1aa48fc4880bb0c9b8a3bf979d3b917e-Abstract.html
posted an update 12 days ago
Post
115
Decades before the modern scaling laws, this paper showed that neural networks behavior under scale follows remarkably predictable laws.
In 1993, researchers at Bell Labs were grappling with a constraint that feels entirely familiar (and contemporary): datasets were outgrowing the available hardware, and training a model to the end was becoming too expensive. To evaluate an architectural tweak to a state-of-the-art model (at the time it was LeNet) on 60,000 samples meant burning up to three weeks of compute time.
To save compute, people would train candidate architectures on small subsets of the data, assuming that the top performer at small scale would remain the top performer at full scale. But with our future wisdom, we know this is not the case.
In "Learning Curves: Asymptotic Values and Rate of Convergence (NeurIPS 93)", using insights from statistical mechanics, they proposed a practical and principled method for predicting the performance of classifiers trained on large datasets (at the time, models were assumed to be large enough). The method was based on a simple power-law modeling of the expected training and test errors.
It is often noted that many of today's breakthroughs in AI and deep learning are actually decades-old concepts that simply lacked the computational power to be tested at the time. While there is some truth to that, it highlights a more valuable lesson: there is immense worth in revisiting early literature and reflecting on foundational ideas we may have prematurely left behind.
So, go explore and find your own inspiration. The current trend has enough champions already!
In 1993, researchers at Bell Labs were grappling with a constraint that feels entirely familiar (and contemporary): datasets were outgrowing the available hardware, and training a model to the end was becoming too expensive. To evaluate an architectural tweak to a state-of-the-art model (at the time it was LeNet) on 60,000 samples meant burning up to three weeks of compute time.
To save compute, people would train candidate architectures on small subsets of the data, assuming that the top performer at small scale would remain the top performer at full scale. But with our future wisdom, we know this is not the case.
In "Learning Curves: Asymptotic Values and Rate of Convergence (NeurIPS 93)", using insights from statistical mechanics, they proposed a practical and principled method for predicting the performance of classifiers trained on large datasets (at the time, models were assumed to be large enough). The method was based on a simple power-law modeling of the expected training and test errors.
It is often noted that many of today's breakthroughs in AI and deep learning are actually decades-old concepts that simply lacked the computational power to be tested at the time. While there is some truth to that, it highlights a more valuable lesson: there is immense worth in revisiting early literature and reflecting on foundational ideas we may have prematurely left behind.
So, go explore and find your own inspiration. The current trend has enough champions already!
- 2 replies
replied to their post 12 days ago
It's a big topic with so many important papers, depending on which aspect you're interested in studying. But you can have a general overview of the current state of the field from this paper, for example: https://arxiv.org/abs/2503.12067
And yeah, I think "Knowledge sharing" or "transfer" are good! But not as charming, like you said :)
reacted to their post with 🧠 16 days ago
Post
284
It has been more than a decade now since the knowledge distillation paper came out.
Knowledge Distillation (KD) is one of my favorite topics, but I have to confess that I'm not a huge fan of the term because I find it confusing (or at least, it has became so over time).
The idea behind KD is not novel; it was there almost a decade before the paper came out (and arguably even a decade before that, back to 1990-91). But this paper is the one that clicked, the one that made the topic much more popular and introduced it to a broader audience.
First, the timing and the authors played a big role: we have Geoffrey Hinton, Oriol Vinyals, and Jeff Dean here. And second, Geoffrey Hinton is really good at idea branding: Model compression?! No, no, no! Let's call it "Knowledge Distillation" and use evocative terms such as "Dark Knowledge" to describe what is being transferred.
It's a great name, but as time has passed, the term became a bit of a relic. KD is no longer solely about compression (KD used to be introduced as a method for model compression, but now model compression is just one application of KD). And the other thing is that the word "distillation" implies some sort of potency here, that the student is somehow more powerful than the teacher, which is not the case (but many counterarguments could be made, for example, more powerful compared to another model trained with no teacher)
Nevertheless, the paper is incredibly well-written, short, and fun to read. It's one of few papers that I read several times. Check it out, and maybe share your thoughts on the topic with us here!
If you had to choose another name for Knowledge Distillation, what would it be?
Knowledge Distillation (KD) is one of my favorite topics, but I have to confess that I'm not a huge fan of the term because I find it confusing (or at least, it has became so over time).
The idea behind KD is not novel; it was there almost a decade before the paper came out (and arguably even a decade before that, back to 1990-91). But this paper is the one that clicked, the one that made the topic much more popular and introduced it to a broader audience.
First, the timing and the authors played a big role: we have Geoffrey Hinton, Oriol Vinyals, and Jeff Dean here. And second, Geoffrey Hinton is really good at idea branding: Model compression?! No, no, no! Let's call it "Knowledge Distillation" and use evocative terms such as "Dark Knowledge" to describe what is being transferred.
It's a great name, but as time has passed, the term became a bit of a relic. KD is no longer solely about compression (KD used to be introduced as a method for model compression, but now model compression is just one application of KD). And the other thing is that the word "distillation" implies some sort of potency here, that the student is somehow more powerful than the teacher, which is not the case (but many counterarguments could be made, for example, more powerful compared to another model trained with no teacher)
Nevertheless, the paper is incredibly well-written, short, and fun to read. It's one of few papers that I read several times. Check it out, and maybe share your thoughts on the topic with us here!
If you had to choose another name for Knowledge Distillation, what would it be?
- 6 replies
replied to their post 16 days ago
Distilling the Knowledge in a Neural Network: https://arxiv.org/abs/1503.02531
posted an update 16 days ago
Post
284
It has been more than a decade now since the knowledge distillation paper came out.
Knowledge Distillation (KD) is one of my favorite topics, but I have to confess that I'm not a huge fan of the term because I find it confusing (or at least, it has became so over time).
The idea behind KD is not novel; it was there almost a decade before the paper came out (and arguably even a decade before that, back to 1990-91). But this paper is the one that clicked, the one that made the topic much more popular and introduced it to a broader audience.
First, the timing and the authors played a big role: we have Geoffrey Hinton, Oriol Vinyals, and Jeff Dean here. And second, Geoffrey Hinton is really good at idea branding: Model compression?! No, no, no! Let's call it "Knowledge Distillation" and use evocative terms such as "Dark Knowledge" to describe what is being transferred.
It's a great name, but as time has passed, the term became a bit of a relic. KD is no longer solely about compression (KD used to be introduced as a method for model compression, but now model compression is just one application of KD). And the other thing is that the word "distillation" implies some sort of potency here, that the student is somehow more powerful than the teacher, which is not the case (but many counterarguments could be made, for example, more powerful compared to another model trained with no teacher)
Nevertheless, the paper is incredibly well-written, short, and fun to read. It's one of few papers that I read several times. Check it out, and maybe share your thoughts on the topic with us here!
If you had to choose another name for Knowledge Distillation, what would it be?
Knowledge Distillation (KD) is one of my favorite topics, but I have to confess that I'm not a huge fan of the term because I find it confusing (or at least, it has became so over time).
The idea behind KD is not novel; it was there almost a decade before the paper came out (and arguably even a decade before that, back to 1990-91). But this paper is the one that clicked, the one that made the topic much more popular and introduced it to a broader audience.
First, the timing and the authors played a big role: we have Geoffrey Hinton, Oriol Vinyals, and Jeff Dean here. And second, Geoffrey Hinton is really good at idea branding: Model compression?! No, no, no! Let's call it "Knowledge Distillation" and use evocative terms such as "Dark Knowledge" to describe what is being transferred.
It's a great name, but as time has passed, the term became a bit of a relic. KD is no longer solely about compression (KD used to be introduced as a method for model compression, but now model compression is just one application of KD). And the other thing is that the word "distillation" implies some sort of potency here, that the student is somehow more powerful than the teacher, which is not the case (but many counterarguments could be made, for example, more powerful compared to another model trained with no teacher)
Nevertheless, the paper is incredibly well-written, short, and fun to read. It's one of few papers that I read several times. Check it out, and maybe share your thoughts on the topic with us here!
If you had to choose another name for Knowledge Distillation, what would it be?
- 6 replies
upvoted a collection 24 days ago
upvoted an article 24 days ago
Article
Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler
- +3
ariG23498, sayakpaul, sergiopaniego, ror, pcuenq
• • 140 reacted to their post with 🚀 24 days ago
Post
2664
What if you could train a model on just 10 images instead of 60,000 and still get close to the same performance?
Traditional machine learning requires thousands, even millions, of data points to achieve high accuracy. But what if we could "distill" the entire dataset into just a few synthetic samples?
This is what Dataset Distillation offers. Unlike traditional knowledge distillation, we keep the model fixed and distill the knowledge contained in a massive training set into a tiny set of synthetic distilled images.
The goal is to train a model on this ultra-small set and achieve performance that almost matches what the same model would get when trained on the massive original dataset.
For example, training on only 10 distilled MNIST images (this is equivalent to a single image per class) yields 94% accuracy, compared to 99% when training on the full 60,000 images.
Interestingly, these distilled images look significantly different (as you can see in the image below) from natural images because they are optimized for model training rather than for matching the correct data distribution.
But that's not all.
Most importantly, this same method opens the door to a potent form of data poisoning. Because distilled images are specifically optimized for rapid learning, an attacker can create a tiny set of adversarial distilled images to cause a well-trained model to forget or misclassify a specific category.
What I find fascinating about dataset distillation is this: it mimics human-like learning by letting a model grasp a concept from a single example, but it does so using alien synthetic images that mean absolutely nothing to a human eye!
What about you? What are your thoughts on it?
Traditional machine learning requires thousands, even millions, of data points to achieve high accuracy. But what if we could "distill" the entire dataset into just a few synthetic samples?
This is what Dataset Distillation offers. Unlike traditional knowledge distillation, we keep the model fixed and distill the knowledge contained in a massive training set into a tiny set of synthetic distilled images.
The goal is to train a model on this ultra-small set and achieve performance that almost matches what the same model would get when trained on the massive original dataset.
For example, training on only 10 distilled MNIST images (this is equivalent to a single image per class) yields 94% accuracy, compared to 99% when training on the full 60,000 images.
Interestingly, these distilled images look significantly different (as you can see in the image below) from natural images because they are optimized for model training rather than for matching the correct data distribution.
But that's not all.
Most importantly, this same method opens the door to a potent form of data poisoning. Because distilled images are specifically optimized for rapid learning, an attacker can create a tiny set of adversarial distilled images to cause a well-trained model to forget or misclassify a specific category.
What I find fascinating about dataset distillation is this: it mimics human-like learning by letting a model grasp a concept from a single example, but it does so using alien synthetic images that mean absolutely nothing to a human eye!
What about you? What are your thoughts on it?
- 2 replies
replied to their post 25 days ago
The Dataset Distillation paper: https://arxiv.org/abs/1811.10959
posted an update 25 days ago
Post
2664
What if you could train a model on just 10 images instead of 60,000 and still get close to the same performance?
Traditional machine learning requires thousands, even millions, of data points to achieve high accuracy. But what if we could "distill" the entire dataset into just a few synthetic samples?
This is what Dataset Distillation offers. Unlike traditional knowledge distillation, we keep the model fixed and distill the knowledge contained in a massive training set into a tiny set of synthetic distilled images.
The goal is to train a model on this ultra-small set and achieve performance that almost matches what the same model would get when trained on the massive original dataset.
For example, training on only 10 distilled MNIST images (this is equivalent to a single image per class) yields 94% accuracy, compared to 99% when training on the full 60,000 images.
Interestingly, these distilled images look significantly different (as you can see in the image below) from natural images because they are optimized for model training rather than for matching the correct data distribution.
But that's not all.
Most importantly, this same method opens the door to a potent form of data poisoning. Because distilled images are specifically optimized for rapid learning, an attacker can create a tiny set of adversarial distilled images to cause a well-trained model to forget or misclassify a specific category.
What I find fascinating about dataset distillation is this: it mimics human-like learning by letting a model grasp a concept from a single example, but it does so using alien synthetic images that mean absolutely nothing to a human eye!
What about you? What are your thoughts on it?
Traditional machine learning requires thousands, even millions, of data points to achieve high accuracy. But what if we could "distill" the entire dataset into just a few synthetic samples?
This is what Dataset Distillation offers. Unlike traditional knowledge distillation, we keep the model fixed and distill the knowledge contained in a massive training set into a tiny set of synthetic distilled images.
The goal is to train a model on this ultra-small set and achieve performance that almost matches what the same model would get when trained on the massive original dataset.
For example, training on only 10 distilled MNIST images (this is equivalent to a single image per class) yields 94% accuracy, compared to 99% when training on the full 60,000 images.
Interestingly, these distilled images look significantly different (as you can see in the image below) from natural images because they are optimized for model training rather than for matching the correct data distribution.
But that's not all.
Most importantly, this same method opens the door to a potent form of data poisoning. Because distilled images are specifically optimized for rapid learning, an attacker can create a tiny set of adversarial distilled images to cause a well-trained model to forget or misclassify a specific category.
What I find fascinating about dataset distillation is this: it mimics human-like learning by letting a model grasp a concept from a single example, but it does so using alien synthetic images that mean absolutely nothing to a human eye!
What about you? What are your thoughts on it?
- 2 replies
replied to their post 30 days ago
Yeah, watched it. Amazing video!
reacted to their post with 🧠 about 1 month ago
Post
4984
It was supposed to be a failed experiment. Instead, it led to the discovery of one of the most intriguing phenomena in neural networks, simply because a researcher forgot to turn it off and left it running....for a week!
In 2022, researchers at OpenAI were studying how neural networks generalize from their training data. For this task, they were training small transformer models to perform modular arithmetic.
The thing is, neural networks are weird. When a model has an abundance of parameters (like neural nets), it can easily overfit. It essentially memorizes its training data, scoring a perfect 100% accuracy when tested on it, but remains completely clueless when faced with any new instances not present in the training set (close to 0 accuracy). It is like memorizing 1 + 2 = 3 without understanding the concept of addition, so if 2 + 3 wasn't in the training set, the model fails miserably!
Usually, when a model overfits like this, people just cut their losses, turn off the experiment, and move on with their lives.
But sometimes they forget. And that is exactly what happened to our researchers at OpenAI. A week later, they checked back in, and a miracle had happened!
They discovered Grokking (And no, this has nothing to do with xAI's Grok , the term was originally coined by sci-fi author Robert Heinlein to mean understanding something so deeply that it becomes part of you). Grokking is when a neural network suddenly and abruptly learns to generalize long after it has overfitted. Just take a look at the graph in the image below!
Spooky, right! I told you neural nets are weird!
In 2022, researchers at OpenAI were studying how neural networks generalize from their training data. For this task, they were training small transformer models to perform modular arithmetic.
The thing is, neural networks are weird. When a model has an abundance of parameters (like neural nets), it can easily overfit. It essentially memorizes its training data, scoring a perfect 100% accuracy when tested on it, but remains completely clueless when faced with any new instances not present in the training set (close to 0 accuracy). It is like memorizing 1 + 2 = 3 without understanding the concept of addition, so if 2 + 3 wasn't in the training set, the model fails miserably!
Usually, when a model overfits like this, people just cut their losses, turn off the experiment, and move on with their lives.
But sometimes they forget. And that is exactly what happened to our researchers at OpenAI. A week later, they checked back in, and a miracle had happened!
They discovered Grokking (And no, this has nothing to do with xAI's Grok , the term was originally coined by sci-fi author Robert Heinlein to mean understanding something so deeply that it becomes part of you). Grokking is when a neural network suddenly and abruptly learns to generalize long after it has overfitted. Just take a look at the graph in the image below!
Spooky, right! I told you neural nets are weird!
- 4 replies
reacted to danielhanchen's post with 🚀 about 1 month ago
Post
4263
Google releases Gemma 4 QAT. ✨
You can now run Gemma 4 at 3x less memory with near original performance.
QAT makes it possible to run Gemma 4 26B-A4B on 16GB RAM.
GGUFs: https://huggingface.co/collections/unsloth/gemma-4-qat
QAT Guide: https://unsloth.ai/docs/models/gemma-4/qat
You can now run Gemma 4 at 3x less memory with near original performance.
QAT makes it possible to run Gemma 4 26B-A4B on 16GB RAM.
GGUFs: https://huggingface.co/collections/unsloth/gemma-4-qat
QAT Guide: https://unsloth.ai/docs/models/gemma-4/qat
- 1 reply
replied to their post about 1 month ago
Read the Grokking paper:
https://huggingface.co/papers/2201.02177
posted an update about 1 month ago
Post
4984
It was supposed to be a failed experiment. Instead, it led to the discovery of one of the most intriguing phenomena in neural networks, simply because a researcher forgot to turn it off and left it running....for a week!
In 2022, researchers at OpenAI were studying how neural networks generalize from their training data. For this task, they were training small transformer models to perform modular arithmetic.
The thing is, neural networks are weird. When a model has an abundance of parameters (like neural nets), it can easily overfit. It essentially memorizes its training data, scoring a perfect 100% accuracy when tested on it, but remains completely clueless when faced with any new instances not present in the training set (close to 0 accuracy). It is like memorizing 1 + 2 = 3 without understanding the concept of addition, so if 2 + 3 wasn't in the training set, the model fails miserably!
Usually, when a model overfits like this, people just cut their losses, turn off the experiment, and move on with their lives.
But sometimes they forget. And that is exactly what happened to our researchers at OpenAI. A week later, they checked back in, and a miracle had happened!
They discovered Grokking (And no, this has nothing to do with xAI's Grok , the term was originally coined by sci-fi author Robert Heinlein to mean understanding something so deeply that it becomes part of you). Grokking is when a neural network suddenly and abruptly learns to generalize long after it has overfitted. Just take a look at the graph in the image below!
Spooky, right! I told you neural nets are weird!
In 2022, researchers at OpenAI were studying how neural networks generalize from their training data. For this task, they were training small transformer models to perform modular arithmetic.
The thing is, neural networks are weird. When a model has an abundance of parameters (like neural nets), it can easily overfit. It essentially memorizes its training data, scoring a perfect 100% accuracy when tested on it, but remains completely clueless when faced with any new instances not present in the training set (close to 0 accuracy). It is like memorizing 1 + 2 = 3 without understanding the concept of addition, so if 2 + 3 wasn't in the training set, the model fails miserably!
Usually, when a model overfits like this, people just cut their losses, turn off the experiment, and move on with their lives.
But sometimes they forget. And that is exactly what happened to our researchers at OpenAI. A week later, they checked back in, and a miracle had happened!
They discovered Grokking (And no, this has nothing to do with xAI's Grok , the term was originally coined by sci-fi author Robert Heinlein to mean understanding something so deeply that it becomes part of you). Grokking is when a neural network suddenly and abruptly learns to generalize long after it has overfitted. Just take a look at the graph in the image below!
Spooky, right! I told you neural nets are weird!
- 4 replies
replied to their post about 1 month ago
posted an update about 1 month ago
Post
1026
Human brains don't recreate every pixel to understand the world!
Most current models in genomics, proteomics, and single-cell transcriptomics rely on generative objectives like masked language modeling or next token prediction. While effective, these architectures waste significant capacity reconstructing raw, noisy sequence details that may not carry functional biological meaning.
But a promising, more efficient alternative is emerging: Joint-Embedding Predictive Architecture (JEPA)
Originally introduced by Yann LeCun for computer vision, JEPA is a non-generative, self-supervised learning (SSL) framework. Instead of predicting raw inputs, it operates as a world model that predicts abstract semantic embeddings in latent space.
Recently, the JEPA framework (and its more efficient LeJEPA variant) has been adapted into the biological sciences to develop performing foundation models and to improve on already existing ones.
It's interesting how each adaptation modified and tailored JEPA to suit its specific biological domain, whether by experimenting with different backbones or complementing the objective with other loss terms.
For example, JEPA-DNA and ProteinJEPA used JEPA as a continual pre-training framework to enhance existing foundation models without training from scratch, while Cell-JEPA and JEPA-DNA employed a hybrid objective that combines the JEPA loss with a traditional language modeling loss.
The article below provides an overview of these implementations, along with others that came out this year. As always, your thoughts and feedback are welcome and highly appreciated!
Link to the article is in the first comment 👇
Most current models in genomics, proteomics, and single-cell transcriptomics rely on generative objectives like masked language modeling or next token prediction. While effective, these architectures waste significant capacity reconstructing raw, noisy sequence details that may not carry functional biological meaning.
But a promising, more efficient alternative is emerging: Joint-Embedding Predictive Architecture (JEPA)
Originally introduced by Yann LeCun for computer vision, JEPA is a non-generative, self-supervised learning (SSL) framework. Instead of predicting raw inputs, it operates as a world model that predicts abstract semantic embeddings in latent space.
Recently, the JEPA framework (and its more efficient LeJEPA variant) has been adapted into the biological sciences to develop performing foundation models and to improve on already existing ones.
It's interesting how each adaptation modified and tailored JEPA to suit its specific biological domain, whether by experimenting with different backbones or complementing the objective with other loss terms.
For example, JEPA-DNA and ProteinJEPA used JEPA as a continual pre-training framework to enhance existing foundation models without training from scratch, while Cell-JEPA and JEPA-DNA employed a hybrid objective that combines the JEPA loss with a traditional language modeling loss.
The article below provides an overview of these implementations, along with others that came out this year. As always, your thoughts and feedback are welcome and highly appreciated!
Link to the article is in the first comment 👇
- 3 replies
upvoted a collection about 1 month ago