@tomwhi.bsky.social
Reposted
"o3, show me a page from the instruction manual for a machine that turns everything into crabs"
Intrigued by the warning it created: "make the warning page of what happens if you insert a crab into the Universal Crabbifier"
"show me a security camera still..."
(all zero shot, first try)
Intrigued by the warning it created: "make the warning page of what happens if you insert a crab into the Universal Crabbifier"
"show me a security camera still..."
(all zero shot, first try)
May 19, 2025 at 8:43 PM
"o3, show me a page from the instruction manual for a machine that turns everything into crabs"
Intrigued by the warning it created: "make the warning page of what happens if you insert a crab into the Universal Crabbifier"
"show me a security camera still..."
(all zero shot, first try)
Intrigued by the warning it created: "make the warning page of what happens if you insert a crab into the Universal Crabbifier"
"show me a security camera still..."
(all zero shot, first try)
Reposted
LLMs may allow computational studies of literature to get beyond word-counting and model things readers value, like surprise. But — what exactly is "surprise"? New work by Annaliese Bissell, Ella Paulin, and @andrewpiper.bsky.social. aclanthology.org/2025.wnu-1.7/
May 10, 2025 at 2:29 PM
LLMs may allow computational studies of literature to get beyond word-counting and model things readers value, like surprise. But — what exactly is "surprise"? New work by Annaliese Bissell, Ella Paulin, and @andrewpiper.bsky.social. aclanthology.org/2025.wnu-1.7/
Reposted
Another one of those little shocking AI moments: this sound clip was generated in 46 seconds on my home PC from the script below. Just the text
Nari Lab's Dia does some of the best expressive AI voice I have seen and it is open weights & created by two undergrads with no funding
Nari Lab's Dia does some of the best expressive AI voice I have seen and it is open weights & created by two undergrads with no funding
April 22, 2025 at 9:12 PM
Another one of those little shocking AI moments: this sound clip was generated in 46 seconds on my home PC from the script below. Just the text
Nari Lab's Dia does some of the best expressive AI voice I have seen and it is open weights & created by two undergrads with no funding
Nari Lab's Dia does some of the best expressive AI voice I have seen and it is open weights & created by two undergrads with no funding
Reposted
Magi-1: The Autoregressive Diffusion Video Generation Model
🥇 The first autoregressive video model with top-tier quality output
🔓 100% open-source & tech report
📊 Exceptional performance on major benchmarks
🥇 The first autoregressive video model with top-tier quality output
🔓 100% open-source & tech report
📊 Exceptional performance on major benchmarks
April 22, 2025 at 6:10 AM
Magi-1: The Autoregressive Diffusion Video Generation Model
🥇 The first autoregressive video model with top-tier quality output
🔓 100% open-source & tech report
📊 Exceptional performance on major benchmarks
🥇 The first autoregressive video model with top-tier quality output
🔓 100% open-source & tech report
📊 Exceptional performance on major benchmarks
Reposted
Comparing Human versus LLM Judges
TREC, which is a community of researchers in information retrieval and natural language processing convened by the NIST, found that an independent human judge correlates better with GPT-4o than a human judge.
TREC, which is a community of researchers in information retrieval and natural language processing convened by the NIST, found that an independent human judge correlates better with GPT-4o than a human judge.
April 22, 2025 at 7:17 PM
Comparing Human versus LLM Judges
TREC, which is a community of researchers in information retrieval and natural language processing convened by the NIST, found that an independent human judge correlates better with GPT-4o than a human judge.
TREC, which is a community of researchers in information retrieval and natural language processing convened by the NIST, found that an independent human judge correlates better with GPT-4o than a human judge.
Reposted
Lvmin Zhang launched FramePack, a novel approach for next - frame prediction models in video generation.
Project: lllyasviel.github.io/frame_pack_g...
Image-to-5-Seconds (30fps, 150 frames)
Project: lllyasviel.github.io/frame_pack_g...
Image-to-5-Seconds (30fps, 150 frames)
April 21, 2025 at 2:39 AM
Lvmin Zhang launched FramePack, a novel approach for next - frame prediction models in video generation.
Project: lllyasviel.github.io/frame_pack_g...
Image-to-5-Seconds (30fps, 150 frames)
Project: lllyasviel.github.io/frame_pack_g...
Image-to-5-Seconds (30fps, 150 frames)
Reposted
Microsoft's BitNet b1.58 2B4T — the first large-scale, native 1-bit LLM🚀🚀
BitNet achieves performance on par with leading full-precision LLMs — and it’s blazingly fast⚡️⚡️uses much lower memory🎉
Everything is open-sourced, per them.
BitNet achieves performance on par with leading full-precision LLMs — and it’s blazingly fast⚡️⚡️uses much lower memory🎉
Everything is open-sourced, per them.
April 16, 2025 at 4:18 AM
Microsoft's BitNet b1.58 2B4T — the first large-scale, native 1-bit LLM🚀🚀
BitNet achieves performance on par with leading full-precision LLMs — and it’s blazingly fast⚡️⚡️uses much lower memory🎉
Everything is open-sourced, per them.
BitNet achieves performance on par with leading full-precision LLMs — and it’s blazingly fast⚡️⚡️uses much lower memory🎉
Everything is open-sourced, per them.
Reposted
A way to help models "be aware of their own capabilities and limitations" from @jacobeisenstein.bsky.social et al: arxiv.org/abs/2503.14481 #MLSky
March 22, 2025 at 4:09 PM
A way to help models "be aware of their own capabilities and limitations" from @jacobeisenstein.bsky.social et al: arxiv.org/abs/2503.14481 #MLSky
Reposted
The Differences between Deep Research, Deep Research, and Deep Research by Han Lee
This blog post examines the various flavors of “Deep Research” from a technical implementation perspective.
leehanchung.github.io/blogs/2025/0...
This blog post examines the various flavors of “Deep Research” from a technical implementation perspective.
leehanchung.github.io/blogs/2025/0...
The Differences between Deep Research, Deep Research, and Deep Research
Dive into the world of AI engineering with this exploration of Deep Research in report generation. Learn how LLM-as-a-judge systems, machine learning, and ev...
leehanchung.github.io
March 19, 2025 at 5:52 AM
The Differences between Deep Research, Deep Research, and Deep Research by Han Lee
This blog post examines the various flavors of “Deep Research” from a technical implementation perspective.
leehanchung.github.io/blogs/2025/0...
This blog post examines the various flavors of “Deep Research” from a technical implementation perspective.
leehanchung.github.io/blogs/2025/0...
Reposted
A very exciting day for open-source AI! We're releasing our biggest open source model yet -- OLMo 2 32B -- and it beats the latest GPT 3.5, GPT 4o mini, and leading open weight models like Qwen and Mistral. As usual, all data, weights, code, etc. are available.
March 13, 2025 at 6:16 PM
A very exciting day for open-source AI! We're releasing our biggest open source model yet -- OLMo 2 32B -- and it beats the latest GPT 3.5, GPT 4o mini, and leading open weight models like Qwen and Mistral. As usual, all data, weights, code, etc. are available.
Reposted
Google has released Gemma 3, the latest open model. Being the most capable and advanced version in the open - source model family, it adds highly requested features like longer context and multimodality.
March 13, 2025 at 5:35 AM
Google has released Gemma 3, the latest open model. Being the most capable and advanced version in the open - source model family, it adds highly requested features like longer context and multimodality.
Reposted
Introducing our Gemma 3 open models, the most capable models that you can run on a single GPU or TPU. Multimodal, multilingual, 128k context length, and exceeds quality of other open models that are an order of magnitude larger in terms of hardware footprint. 🎉
blog.google/technology/d...
blog.google/technology/d...
Introducing Gemma 3: The most capable model you can run on a single GPU or TPU
Today, we're introducing Gemma 3, our most capable, portable and responsible open model yet.
blog.google
March 13, 2025 at 2:55 PM
Introducing our Gemma 3 open models, the most capable models that you can run on a single GPU or TPU. Multimodal, multilingual, 128k context length, and exceeds quality of other open models that are an order of magnitude larger in terms of hardware footprint. 🎉
blog.google/technology/d...
blog.google/technology/d...
Reposted
PapersChat – Chat with Research Papers
PapersChat provides an agentic AI interface for querying papers, retrieving insights from ArXiv & PubMed, and structuring responses efficiently.
github.com/AstraBert/Pa...
PapersChat provides an agentic AI interface for querying papers, retrieving insights from ArXiv & PubMed, and structuring responses efficiently.
github.com/AstraBert/Pa...
March 10, 2025 at 4:47 AM
PapersChat – Chat with Research Papers
PapersChat provides an agentic AI interface for querying papers, retrieving insights from ArXiv & PubMed, and structuring responses efficiently.
github.com/AstraBert/Pa...
PapersChat provides an agentic AI interface for querying papers, retrieving insights from ArXiv & PubMed, and structuring responses efficiently.
github.com/AstraBert/Pa...
Reposted
Mistral has just released a new OCR model, Mistral-OCR. Yet still the usual VLM curse: with challenging manuscripts, it hallucinates completely.
March 6, 2025 at 7:02 PM
Mistral has just released a new OCR model, Mistral-OCR. Yet still the usual VLM curse: with challenging manuscripts, it hallucinates completely.
Reposted
Reposted
An Overview of Large Language Models for Statisticians
- Stat for LLM: How statistical methods can improve LLM uncertainty quantification, interpretability, trustworthiness & more.
- Stat for LLM: How statistical methods can improve LLM uncertainty quantification, interpretability, trustworthiness & more.
February 27, 2025 at 4:09 AM
An Overview of Large Language Models for Statisticians
- Stat for LLM: How statistical methods can improve LLM uncertainty quantification, interpretability, trustworthiness & more.
- Stat for LLM: How statistical methods can improve LLM uncertainty quantification, interpretability, trustworthiness & more.
Reposted
I asked Claude “Make an interactive artifact that will illustrate to me why I should not start Civ VII right now.”
This is what it came up with on its own.
This is what it came up with on its own.
February 8, 2025 at 10:34 PM
I asked Claude “Make an interactive artifact that will illustrate to me why I should not start Civ VII right now.”
This is what it came up with on its own.
This is what it came up with on its own.
Reposted
Why do LLMs trained on over 90% English text perform so well in non-English languages?
They find that they learn to share highly abstract grammatical concept representations, even across unrelated languages!
They find that they learn to share highly abstract grammatical concept representations, even across unrelated languages!
February 6, 2025 at 7:19 AM
Why do LLMs trained on over 90% English text perform so well in non-English languages?
They find that they learn to share highly abstract grammatical concept representations, even across unrelated languages!
They find that they learn to share highly abstract grammatical concept representations, even across unrelated languages!
Reposted
Fellow journos covering "AI": Please don't do their PR for them! "Virtual employees" is a harmful anthropomorphism in that it (a) is false; (b) confuses readers about the emerging technology and inaccurately lends human attributes like agency, accountability, etc.; and (c) harms humans in real jobs.
‘Virtual employees’ could join workforce as soon as this year, OpenAI boss says
‘Virtual employees’ could join workforce as soon as this year, OpenAI boss says
Sam Altman says tools that carry out jobs anonymously, known as AI agents, could transform business output
Virtual employees could join workforces this year and transform how companies work, according to the chief executive of OpenAI.
The first…
www.theguardian.com
January 6, 2025 at 5:07 PM
Fellow journos covering "AI": Please don't do their PR for them! "Virtual employees" is a harmful anthropomorphism in that it (a) is false; (b) confuses readers about the emerging technology and inaccurately lends human attributes like agency, accountability, etc.; and (c) harms humans in real jobs.
Reposted
Yes, DeepSeek R1's release is impressive. But the real story is what happened in just 7 days after:
Original release: 8 models, 540K downloads. Just the beginning...
The community turned those open-weight models into +550 NEW models on @huggingface. Total downloads? 2.5M—nearly 5X the originals.
Original release: 8 models, 540K downloads. Just the beginning...
The community turned those open-weight models into +550 NEW models on @huggingface. Total downloads? 2.5M—nearly 5X the originals.
January 27, 2025 at 4:24 PM
Yes, DeepSeek R1's release is impressive. But the real story is what happened in just 7 days after:
Original release: 8 models, 540K downloads. Just the beginning...
The community turned those open-weight models into +550 NEW models on @huggingface. Total downloads? 2.5M—nearly 5X the originals.
Original release: 8 models, 540K downloads. Just the beginning...
The community turned those open-weight models into +550 NEW models on @huggingface. Total downloads? 2.5M—nearly 5X the originals.
Reposted
We’re excited to introduce Transformer², a machine learning system that dynamically adjusts its weights for various tasks!
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
January 15, 2025 at 5:49 AM
We’re excited to introduce Transformer², a machine learning system that dynamically adjusts its weights for various tasks!
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
Yet more interesting research by sakana.ai
We’re excited to introduce Transformer², a machine learning system that dynamically adjusts its weights for various tasks!
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
sakana.ai/transformer-...
Adaptation is a remarkable natural phenomenon, like how the octopus blends into its environment, or how the brain rewires itself after injury.
🧵 1/N
January 26, 2025 at 10:57 AM
Yet more interesting research by sakana.ai
Reposted
ByteDance Doubao-1.5-pro
- Includes a "Deep Thinking" mode, surpassing O1-preview and O1 models on the AIME benchmark.
- Outperforms deepseek-v3, gpt4o, and llama3.1-405B on popular benchmarks.
team.doubao.com/en/special/d...
- Includes a "Deep Thinking" mode, surpassing O1-preview and O1 models on the AIME benchmark.
- Outperforms deepseek-v3, gpt4o, and llama3.1-405B on popular benchmarks.
team.doubao.com/en/special/d...
January 24, 2025 at 9:43 PM
ByteDance Doubao-1.5-pro
- Includes a "Deep Thinking" mode, surpassing O1-preview and O1 models on the AIME benchmark.
- Outperforms deepseek-v3, gpt4o, and llama3.1-405B on popular benchmarks.
team.doubao.com/en/special/d...
- Includes a "Deep Thinking" mode, surpassing O1-preview and O1 models on the AIME benchmark.
- Outperforms deepseek-v3, gpt4o, and llama3.1-405B on popular benchmarks.
team.doubao.com/en/special/d...