Picture

Wei Shi @weishi

Central Region, Singapore Joined November 2008
  • Tweets
    59
  • Followers
    83
  • Following
    929
  • Likes
    3
Wei Shi retweeted
timelessdev Profile Picture

Jacek (Jomsborg.eth) @timelessdev

4 days ago

@bllchmbrs @ApacheSpark @matei_zaharia @alighodsi True. Its just begining. I try to learn DSPy by example: RAG fueled by DSPy medium.com/@JacekWo/rag-f…

Wei Shi retweeted
0xdiicell Profile Picture

Diicell @0xdiicell

2 months ago

Excited to share new RAG Demo Application Template! code: github.com/diicellman/dsp… Built with @FastAPI & @Gradio, powered by @stanfordnlp DSPy, and made fully local with @ollama. It's an example for devs exploring DSPy, RAG, or creating AI-driven apps locally.

Wei Shi retweeted
_akhaliq Profile Picture

AK @_akhaliq

2 days ago

Meta presents AdvPrompter Fast Adaptive Adversarial Prompting for LLMs While recently Large Language Models (LLMs) have achieved remarkable successes, they are vulnerable to certain jailbreaking attacks that lead to generation of inappropriate or harmful content.

Wei Shi retweeted
algo_diver Profile Picture

chansung @algo_diver

2 days ago

The best CUDA intro course by @nvidia with 460 bite sized videos. It was the course released with Udacity 9 yrs ago. It is kinda old, but you can grasp core ideas around it. youtube.com/playlist?list=…

Wei Shi retweeted
iScienceLuvr Profile Picture

Tanishq Mathew Abraham, Ph.D. @iScienceLuvr

3 days ago

A reminder that most evaluation benchmarks are garbage

typedfemale Profile Picture

typedfemale @typedfemale

4 days ago

A reminder that most evaluation benchmarks are garbage

Wei Shi retweeted
p_nawrot Profile Picture

Piotr Nawrot @p_nawrot

4 days ago

Two free medium-compute Mixture-Of-Experts research ideas: Prerequisite: Mixtral 8x7B is 32 layers, at each layer there are 8 experts, each token is assigned to 2 experts at a given layer. 1) Dynamic Expert Assignment in MoE Models Every token is assigned to 2*32=64 experts in…

Wei Shi retweeted
bllchmbrs Profile Picture

Bill Chambers @bllchmbrs

5 days ago

DSPy is an amazing tool because it gives engineers the power to move from model provider to model provider in. a PRINCIPLED WAY. Engineers just don't know it yet. I want to change that.

Wei Shi retweeted
gowthami_s Profile Picture

Gowthami Somepalli @gowthami_s

4 days ago

✨ Excited ✨ Wanna get a whirlwind tour of memorization in diffusion models, how to find it, how to mitigate it, drop by the talk! Will discuss all my 3 papers (including style memorization) on this topic. Papers: 1. arxiv.org/abs/2212.03860 (CVPR'23) 2.…

CohereForAI Profile Picture

Cohere For AI @CohereForAI

4 days ago

✨ Excited ✨ Wanna get a whirlwind tour of memorization in diffusion models, how to find it, how to mitigate it, drop by the talk! Will discuss all my 3 papers (including style memorization) on this topic. Papers: 1. arxiv.org/abs/2212.03860 (CVPR'23) 2.…

Wei Shi retweeted
lateinteraction Profile Picture

Omar Khattab @lateinteraction

4 days ago

Why are Signatures a thing in DSPy? Standard prompts conflate interface (“what should the LM do?”) with implementation (“how do we tell it to do that?”). DSPy signatures isolate the former so we can infer & learn the latter from data — in the context of a bigger program.

bllchmbrs Profile Picture

Bill Chambers @bllchmbrs

4 days ago

Why are Signatures a thing in DSPy? Standard prompts conflate interface (“what should the LM do?”) with implementation (“how do we tell it to do that?”). DSPy signatures isolate the former so we can infer & learn the latter from data — in the context of a bigger program.

Wei Shi retweeted
omarsar0 Profile Picture

elvis @omarsar0

4 days ago

Make Your LLM Fully Utilize the Context Microsoft researchers and collaborators present an approach to overcome the lost-in-the-middle challenge common in LLMs. It applies an explicit "information-intensive" training procedure on Mistral-7B to enable the LLM to fully utilize…

Wei Shi retweeted
gneubig Profile Picture

Graham Neubig @gneubig

5 days ago

We're having a big event on agents at CMU on May 2-3 (one week from now), all are welcome! cmu-agent-workshop.github.io It will feature: * Invited talks from @alsuhr @ysu_nlp @xinyun_chen_ @MaartenSap and @chris_j_paxton * Posters of cutting edge research * Seminars and hackathons

frankxu2004 Profile Picture

Frank Xu @frankxu2004

a week ago

We're having a big event on agents at CMU on May 2-3 (one week from now), all are welcome! cmu-agent-workshop.github.io It will feature: * Invited talks from @alsuhr @ysu_nlp @xinyun_chen_ @MaartenSap and @chris_j_paxton * Posters of cutting edge research * Seminars and hackathons

Wei Shi retweeted
LangChainAI Profile Picture

LangChain @LangChainAI

5 days ago

🤖🏆LangGraph: Can Language Models Solve Olympiad Programming? 🤖🏆 Last week, Princeton researchers released the USACO benchmark dataset and showed that a zero-shot GPT-4 agent only passes 8.7% of the questions. We've implemented this paper in LangGraph and created a tutorial…

Wei Shi retweeted
omarsar0 Profile Picture

elvis @omarsar0

2 weeks ago

Emerging AI Agent Architectures Researchers from IBM and Microsoft present this concise summary of emerging AI agent architectures. It focuses the discussion on capabilities like reasoning, planning, and tool calling which are all needed to build complex AI-powered agentic…

Wei Shi retweeted
dair_ai Profile Picture

DAIR.AI @dair_ai

a week ago

5/ A Survey on Retrieval-Augmented Text Generation for LLMs - presents a comprehensive overview of the RAG domain, its evolution, and challenges. x.com/omarsar0/statu…

omarsar0 Profile Picture

elvis @omarsar0

2 weeks ago

5/ A Survey on Retrieval-Augmented Text Generation for LLMs - presents a comprehensive overview of the RAG domain, its evolution, and challenges. x.com/omarsar0/statu…

Wei Shi retweeted
yoavgo Profile Picture

(((ل()(ل() 'yoav))))👾 @yoavgo

2 weeks ago

We've been looking quite a bit lately at measuring the language-understanding ability of LLMs. This is an arxiv preprint on this line of work (w/ @rtsarfaty @VikaBsmv). arxiv.org/abs/2404.06283

Wei Shi retweeted
omarsar0 Profile Picture

elvis @omarsar0

2 weeks ago

How Faithful are RAG Models? This new paper aims to quantify the tug-of-war between RAG and LLMs' internal prior. It focuses on GPT-4 and other LLMs on question answering for the analysis. It finds that providing correct retrieved information fixes most of the model…

Wei Shi retweeted
reach_vb Profile Picture

Vaibhav (VB) Srivastav @reach_vb

6 days ago

Snowflake dropped a 408B Dense + Hybrid MoE 🔥 > 17B active parameters > 128 experts > trained on 3.5T tokens > uses top-2 gating > fully apache 2.0 licensed (along with data recipe too) > excels at tasks like SQL generation, coding, instruction following > 4K context window,…

Wei Shi retweeted
CShorten30 Profile Picture

Connor Shorten @CShorten30

6 days ago

As you are graduating from ideas to engineering, one of the key concepts to be aware of is Parallel Computing and Concurrency Control. 🔒🔓 I am SUPER excited to share our 94th Weaviate podcast with Magdalen Dobson Manohar! Magdalen is one of the most impressive scientists I…

Wei Shi retweeted
IntuitMachine Profile Picture

Carlos E. Perez @IntuitMachine

a week ago

1/n Learning to Search: How LLMs Can Master Problem-Solving The ability to plan, strategize, and search for solutions lies at the heart of intelligent behavior. While recent advancements in large language models (LLMs) have demonstrated impressive capabilities in various tasks,…

Wei Shi retweeted
deedydas Profile Picture

Deedy @deedydas

3 weeks ago

Growing up, Paul Graham’s principles of using less words gave me a lot of solace. In a British colony like India at a Catholic school, we were taught the opposite: Use fancy words. Prefer verbosity. Signal your class. This distracted so many people from: Just. Knowing. Stuff.

Trends for United States

Kendrick
711 B posts
Bayern
324 B posts
Euphoria
279 B posts
Jack Flaherty
1.952 posts
$VSG
1.677 posts
Vini
102 B posts
Kdot
15,3 B posts
Kroos
148 B posts
#UnderdogPicks
5.995 posts
Aubrey
33,3 B posts
Push Ups
27,5 B posts
Sane
77,2 B posts
Musiala
40,1 B posts
Barron
31,9 B posts
Gillie
6.410 posts
#FlyToPixarFest
Tuchel
31,3 B posts
Bernabeu
24,2 B posts
Kim Min Jae
10,3 B posts
J Cole
38,3 B posts

You might like

Neuromorphics @Intel | MLPerf, TPU perf, Chrome GPU @Google | Android GPU, ASIC design @Nvidia | Robots @MediaLab | HT patch clamp @Tecella | Runner
Brian Anderson
@braindersnn
542 Followers
aliya
@ALIYAHAJI5
196 Followers
Seeking rational truth in a tribal world. Founder of Seldon Capital. Investing in the bottlenecks of human productivity. Formerly Soros Fund Management.
Matthew Fong
@mattfong_
340 Followers
Principal Engineer @CerebrasSystems. PhD in AI, Satisfiability and Constraint Satisfaction Problems. Tweets don't necessarily represent my employer's opinions.
Ozan Erdem
@ozanerdem
624 Followers
Building ML Infra @Netflix. Formerly @alchemyplatform, @DownToLunch, @wandlabs, @twitter. Might be the lead singer of Staind. He/Him
Aaron Lewis
@aaronlewism
2K Followers