Nikhil Soni
@nikhilsoni.bsky.social
ML @ Apple. Focused on Simulation, LLM agents and Cognitive architectures.
Great presentation by Ilya Sutskever yesterday at the #NeurIPS2024 test of time awards! Loved his format of reusing slides from 2014 and talking about what they got right and wrong. He ended with some strong claims about pre-training and predictions about what comes next.
December 14, 2024 at 12:48 PM
Great presentation by Ilya Sutskever yesterday at the #NeurIPS2024 test of time awards! Loved his format of reusing slides from 2014 and talking about what they got right and wrong. He ended with some strong claims about pre-training and predictions about what comes next.
Reposted by Nikhil Soni
Hi NeurIPS!
Explore ~4,500 NeurIPS papers in this interactive visualization:
jalammar.github.io/assets/neuri...
(Click on a point to see the paper on the website)
Uses @cohere.com models and @lelandmcinnes.bsky.social's datamapplot/umap to help make sense of the overwhelming scale of NeurIPS.
Explore ~4,500 NeurIPS papers in this interactive visualization:
jalammar.github.io/assets/neuri...
(Click on a point to see the paper on the website)
Uses @cohere.com models and @lelandmcinnes.bsky.social's datamapplot/umap to help make sense of the overwhelming scale of NeurIPS.
December 10, 2024 at 10:49 PM
Hi NeurIPS!
Explore ~4,500 NeurIPS papers in this interactive visualization:
jalammar.github.io/assets/neuri...
(Click on a point to see the paper on the website)
Uses @cohere.com models and @lelandmcinnes.bsky.social's datamapplot/umap to help make sense of the overwhelming scale of NeurIPS.
Explore ~4,500 NeurIPS papers in this interactive visualization:
jalammar.github.io/assets/neuri...
(Click on a point to see the paper on the website)
Uses @cohere.com models and @lelandmcinnes.bsky.social's datamapplot/umap to help make sense of the overwhelming scale of NeurIPS.
I am in Vancouver for #NeurIPS2024!
At present, my most pressing interests lie in the domains of:
- Social modeling utilizing Multi Agent Systems within simulation
- World simulators
I would welcome the opportunity to engage in a discussion on any of these topics (or any other related subjects).
At present, my most pressing interests lie in the domains of:
- Social modeling utilizing Multi Agent Systems within simulation
- World simulators
I would welcome the opportunity to engage in a discussion on any of these topics (or any other related subjects).
December 10, 2024 at 3:46 AM
I am in Vancouver for #NeurIPS2024!
At present, my most pressing interests lie in the domains of:
- Social modeling utilizing Multi Agent Systems within simulation
- World simulators
I would welcome the opportunity to engage in a discussion on any of these topics (or any other related subjects).
At present, my most pressing interests lie in the domains of:
- Social modeling utilizing Multi Agent Systems within simulation
- World simulators
I would welcome the opportunity to engage in a discussion on any of these topics (or any other related subjects).
Reposted by Nikhil Soni
PydanticAI is here!
An Agent Framework designed for production, from the team who created and maintain @pydantic.bsky.social.
As some of you will know, I've been working on this for some time, can't wait to see what people build with it.
ai.pydantic.dev
An Agent Framework designed for production, from the team who created and maintain @pydantic.bsky.social.
As some of you will know, I've been working on this for some time, can't wait to see what people build with it.
ai.pydantic.dev
Introduction
Agent Framework / shim to use Pydantic with LLMs
ai.pydantic.dev
December 2, 2024 at 11:05 AM
PydanticAI is here!
An Agent Framework designed for production, from the team who created and maintain @pydantic.bsky.social.
As some of you will know, I've been working on this for some time, can't wait to see what people build with it.
ai.pydantic.dev
An Agent Framework designed for production, from the team who created and maintain @pydantic.bsky.social.
As some of you will know, I've been working on this for some time, can't wait to see what people build with it.
ai.pydantic.dev
Reposted by Nikhil Soni
ICLR is a top ML conference. All 10k+ papers from 2025 are in Open Review.
The top rated papers include:
— Scaling LLM interpretability to GPT4 scale
— Changing light source w consistent image
— 100x faster diffusion models
— A provable theory for LLM jailbreaks
Thread...
The top rated papers include:
— Scaling LLM interpretability to GPT4 scale
— Changing light source w consistent image
— 100x faster diffusion models
— A provable theory for LLM jailbreaks
Thread...
November 29, 2024 at 3:16 PM
ICLR is a top ML conference. All 10k+ papers from 2025 are in Open Review.
The top rated papers include:
— Scaling LLM interpretability to GPT4 scale
— Changing light source w consistent image
— 100x faster diffusion models
— A provable theory for LLM jailbreaks
Thread...
The top rated papers include:
— Scaling LLM interpretability to GPT4 scale
— Changing light source w consistent image
— 100x faster diffusion models
— A provable theory for LLM jailbreaks
Thread...
Reposted by Nikhil Soni
Adding my love letter to
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org
November 23, 2024 at 1:40 PM
Adding my love letter to
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.