Karan Goel @krandiash Twitter profile

Pinned Tweet

Karan Goel

@krandiash

4 years

This is basically machine learning

Massimo

@Rainmaker1973

4 years

The story of Nigel Richards, the man from New Zealand who memorized every French word in the French scrabble dictionary and won the French Scrabble Championship without speaking any French

196

3K

16K

60

2K

12K

Last Seen Profiles

@Effie5772

@PlastererMatt

@_frxddyy

@DrAnchalaSobrin

@Teejigl9Ax

@Mikoladeimon

@Snesdude2

@GeorgeBentham7

@Mojtabapacino

@stwmaniax

@OvidElsieSports

@HjortLaris79653

@y6x77

@ero_fanzadmm

@TB_Choi12

@1Majeed21

@KganyaDragon

@MillieBles11311

@heezy_lee

@benqiik

@TindAIR_EU

@1mZhjAT1espA5Uk

@VanderEsch38

@saabvrienden

@bokeplokalmalam

@weizhiai

@deepsixersbr

@jwe1022

@j_igz

@karechan3840

@HardensBites

@TerryAilsa

@gloryzonemusic

@adeliz99

@PankowerFrauen

@NadiaForde

Karan Goel

@krandiash

4 months

got my PhD

43

4

688

Karan Goel

@krandiash

1 year

Successfully defended my PhD yesterday, one of the most fun experiences of my life (barring Covid) thanks to @HazyResearch Time for more fun stuff

40

11

440

Karan Goel

@krandiash

4 years

If you’ve ever wanted to take a grubby Python project and turn it into something that looks more like a well run open-source project (👋 ML researchers), here’s a guide I wrote on how to do it. I was frustrated after Googling for hours, so hope it helps!

Terse Guide to Python Projects | Notion

🤵 by Karan Goel (kgoel [at] cs.stanford.edu) | twitter: @krandiash

krandi.notion.site

11

67

419

Karan Goel

@krandiash

4 months

Tech report coming soon! SSMs are an amazing fit for audio, perplexity numbers with our new architecture blow Transformer baselines out of the water, look at this giant gap on training loss

combin8

@combin8or

4 months

@krandiash @reach_vb SSMs seem like a great fit for TTS. Any details on the model?

0

4

19

45

405

Karan Goel

@krandiash

4 years

🚀Excited to release Robustness Gym, a new Python evaluation toolkit for evaluating the robustness of NLP models, as part of a collaboration between Stanford, Salesforce Research and UNC Chapel-Hill. Paper: Code: pip install!

GitHub - robustness-gym/robustness-gym: Robustness Gym is an evaluation toolkit for machine...

Robustness Gym is an evaluation toolkit for machine learning. - robustness-gym/robustness-gym

github.com

1

82

285

Karan Goel

@krandiash

4 years

(Thread) I finally got GPT-3 access last week (shout out to @gdb ), and took a stab at an experiment that I've been curious about for a while. TLDR: training a model on a dataset entirely generated by GPT-3. You can read my blog at .

GPT3 → Dataset → Task Model? | Notion

TLDR: I go from wanting a machine learning model to getting that trained model, without actually having a dataset.

krandi.notion.site

6

55

260

Karan Goel

@krandiash

2 years

Our work on S4 received a best paper hon. mention at ICLR 🎊

4

10

249

Karan Goel

@krandiash

26 days

congratulations to my friend, cofounder and research buddy @_albertgu on his amazing impact in the world of AI

TIME

@TIME

26 days

TIME's new cover: The 100 most influential people in AI

4K

773

4K

5

11

227

Karan Goel

@krandiash

3 years

Our new preprint is out! 🍣 SaShiMi is a new architecture for modeling raw audio waveforms, built around state-space models like S4. 📜 ⭐️ 🔊 🧵 below

3

48

228

Karan Goel

@krandiash

4 months

Incredibly excited to be releasing our first model, @cartesia_ai Sonic today. Sonic is a voice model based on a new state space model architecture we've developed that's blazing fast, efficient and high quality. It's the first of many models we're building to bring cheap

Cartesia

@cartesia_ai

4 months

Today, we’re excited to release the first step in our mission to build real time multimodal intelligence for every device: Sonic, a blazing fast (🚀 135ms model latency), lifelike generative voice model and API. Read and try Sonic

43

163

821

19

18

214

Karan Goel

@krandiash

2 years

We built an interactive data frame powered by foundation models that can wrangle your unstructured data (images, videos, text docs...) Introducing 🔮 Meerkat! 📃 💻 🌐

3

60

206

Karan Goel

@krandiash

2 years

There’s a weird dichotomy where all the AI researchers I interact with think there’s a lot left to do on designing new architectures that improve over Transformers — but everyone else seems to be entirely unaware that this is even a possibility left to consider

11

5

186

Karan Goel

@krandiash

6 years

Really cool work from the Brunskill lab at Stanford on using model-based RL!

0

57

184

Karan Goel

@krandiash

3 months

we're hiring research engineers / modelers to accelerate model development ship multimodal models in a fast-paced team that's moulding the future of real-time architectures @_albertgu will give you your company sponsored pet yoshi himself

Research Engineer

ABOUT CARTESIA Our mission is to build the next generation of AI: ubiquitous, interactive intelligence that runs wherever you are. Today, not even the best models can continuously process and reason...

jobs.ashbyhq.com

2

11

168

Karan Goel

@krandiash

25 days

2.5 months ago @elevenlabsio put up this comparison with our 10 day old Sonic model: The team took it as a challenge, here's our new scorecard. Higher quality, cheaper & the fastest voice model period. Next 3 months will be fun.

10

24

160

Karan Goel

@krandiash

4 years

Preprint alert! "Model Patching: Closing the Subgroup Performance Gap with Data Augmentation" is now on arXiv! 📑Paper: 🧑‍💻Code: 📹Video: ✍️Blog: Read on to learn more (1/9)

3

37

154

Karan Goel

@krandiash

4 years

Writing a rebuttal for NeurIPS, What I want to say 😏 “Your review is $%*€¥. Try again. 2/10.” What I actually say 😒 “Thanks for the helpful feedback. Your wisdom and insight are truly wondrous and move my soul. I was touched that you think we don’t have enough baselines...”

1

6

132

Karan Goel

@krandiash

2 months

insanity from the research team, Sonic now the only model in production for generative voice breaking 2 digits on latency 🤯

Cartesia

@cartesia_ai

2 months

We're faster now.

7

90

13

2

119

Karan Goel

@krandiash

3 years

Excited to release a new resource for Data Centric AI: ...with a great post by @HazyResearch about our lab's journey in this: This is already a community effort with 20+ folks who have contributed discussion. Please send us PRs!

GitHub - HazyResearch/data-centric-ai: Resources for Data Centric AI

Resources for Data Centric AI. Contribute to HazyResearch/data-centric-ai development by creating an account on GitHub.

github.com

1

35

113

Karan Goel

@krandiash

3 years

Excited to release Meerkat, a new data library for interactive machine learning! We've ( @jundesai , @EyubogluSabri , @HazyResearch ) been building this up over the last couple of months. Read our blog post to learn more:

Meerkat: DataPanels for Machine Learning | Notion

Blog by Sabri Eyuboglu ([email protected]), Arjun Desai ([email protected]), Karan Goel ([email protected])

sabrieyuboglu.notion.site

0

33

102

Karan Goel

@krandiash

4 years

Awesome to see that our MLSys seminar series now has 3k subs on YouTube (and counting): I’m constantly amazed by how many folks I interact with have watched, thanks for tuning in! (and subscribe) @realDanFu @w4nderlus7 @matei_zaharia @HazyResearch

Stanford MLSys Seminars

Stanford seminar on the frontier of machine learning and systems. Live every Thursday, 1:15-3 pm PT.

www.youtube.com

1

19

100

Karan Goel

@krandiash

2 years

Accepted to ICML ‘22 as a long talk!! Shout out and thanks to my brilliant coauthors, @_albertgu @chrisdonahuey @HazyResearch

Karan Goel

@krandiash

3 years

Our new preprint is out! 🍣 SaShiMi is a new architecture for modeling raw audio waveforms, built around state-space models like S4. 📜 ⭐️ 🔊 🧵 below

3

48

228

1

12

98

Karan Goel

@krandiash

4 years

People from the past keep stealing my ideas.

3

2

89

Karan Goel

@krandiash

4 months

we'll be shipping another @cartesia_ai model next week start working with us asap if you want to get early access to all the cool stuff that's coming, this team is 🚢 new models every 2-3 weeks

8

5

89

Karan Goel

@krandiash

4 years

A while back, I wrote a Python library for handling YAML-based configuration in my ML projects. I've been installing (`pip install quinine`) it for my own projects for a while, now you can use it too README:

GitHub - krandiash/quinine: A library to create and manage configuration files, especially for...

A library to create and manage configuration files, especially for machine learning projects. - krandiash/quinine

github.com

4

9

84

Karan Goel

@krandiash

4 months

I'll be at Founders You Should Know in SF tonight to talk about @cartesia_ai , find me if you'd like to chat We're hiring!

4

2

83

Karan Goel

@krandiash

3 years

Really chuffed to see that we've crossed 5000 subs on our MLSys Seminar YouTube after 34 weeks of streaming (). A big thanks to all our speakers and viewers, and the cast ( @realDanFu , @w4nderlus7 , @HazyResearch , @matei_zaharia , Fiodar)!

Stanford MLSys Seminars

Stanford seminar on the frontier of machine learning and systems. Live every Thursday, 1:15-3 pm PT.

www.youtube.com

1

10

81

Karan Goel

@krandiash

3 years

Want to use state space models (S4 -- ) and don't know where to start? We just put up an example script () on how to build a simple S4 model backbone that crosses the previous SOTA on sequential CIFAR (81%) in 30 minutes on a V100!

3

11

79

Karan Goel

@krandiash

3 years

Happy news -- we completed a year streaming the MLSys seminar this week: 42 episodes in 52 weeks! Fun fact: we've crossed 10k watch hours (mindboggling to me), thanks for tuning in! @realDanFu @fiodarkaz @w4nderlus7 @matei_zaharia @HazyResearch

1

9

76

Karan Goel

@krandiash

2 years

Came all the way to neurips to meet Bay Area people here

3

1

77

Karan Goel

@krandiash

10 months

Excited about this incredible SSM from @_albertgu and @tri_dao , and excited to be working with @_albertgu on scaling SSMs at @cartesia_ai . Stay tuned for more.

Albert Gu

@_albertgu

10 months

Quadratic attention has been indispensable for information-dense modalities such as language... until now. Announcing Mamba: a new SSM arch. that has linear-time scaling, ultra long context, and most importantly--outperforms Transformers everywhere we've tried. With @tri_dao 1/

54

418

2K

1

2

74

Karan Goel

@krandiash

2 months

one of the best pieces of advice i ever got (in the context of going to ai conferences) was to spend time with your peers rather than chasing after senior or famous researchers you have more fun, grow together and who knows, maybe some of your peers will be famous one day

1

2

72

Karan Goel

@krandiash

1 month

our 3 part on-device release today with edge, rene and sonic on-device is now out edge is our new open-source library for on-device SSM deployments with new kernels & models this starts our journey to build truly efficient human-like AI that's detethered from the data center

Cartesia

@cartesia_ai

1 month

Today, we’re unveiling a significant milestone in our journey toward ubiquitous artificial intelligence: AI On-Device. Our team pioneered a radically more efficient architecture for AI with state space models (SSMs). Now, we’ve optimized and deployed them at the edge. We believe

11

84

365

8

5

70

Karan Goel

@krandiash

4 months

mamba-2 is here 👀 if you want to work on bleeding edge ssms with a world class research team led by @_albertgu , we're hiring @cartesia_ai

Albert Gu

@_albertgu

4 months

excited to finally release Mamba-2!! 8x larger states, 50% faster training, and even more S's 🐍🐍 Mamba-2 aims to advance the theory of sequence models, developing a framework of connections between SSMs and (linear) attention that we call state space duality (SSD) w/ @tri_dao

11

187

1K

1

6

67

Karan Goel

@krandiash

4 years

Left: Cherry-picked GAN pix in the paper. Right: Me after running the authors’ code from Github and seeing the outputs.

0

4

63

Karan Goel

@krandiash

4 years

Indian society is cursed. The trope of the “qualified woman” whose sole purpose is marriage is frankly infuriating. These idiotic “traditions” permeate even the most liberal parts of India. If you’re Indian, your family probably has people who clutch onto these ideals.

2

6

59

Karan Goel

@krandiash

1 year

We built a data exploration dashboard that we shipped with @togethercompute 's new Red Pajama LLM data release! We embedded the entire Github subset of Red Pajama (releasing indexes + embeddings soon!). Built in 100 lines of Python with @MeerkatML 🚀

3

5

56

Karan Goel

@krandiash

2 months

the amazing thing about building a business in 🇺🇸 your multilingual models are evaluated by native speakers that sit right next to you (we’ve got 10 languages covered)

2

53

Karan Goel

@krandiash

3 months

the team's been busy 2 model updates, 3 new features shipping in the next week you'll be able to control voice emotion via our API soon

2

0

53

Karan Goel

@krandiash

4 months

SSMs are coming

Bryan Catanzaro

@ctnzr

4 months

A 8B-3.5T hybrid SSM model gets better accuracy than an 8B-3.5T transformer trained on the same dataset: * 7% attention, the rest is Mamba2 * MMLU jumps from 50 to 53.6% * Training efficiency is the same * Inference cost is much less

18

77

450

1

3

53

Karan Goel

@krandiash

21 days

we're now organizing some incredible efforts to push forward innovation on model architectures (we'll announce more on this soon) how do we compress a decade of model architecture progress into a year? if you're excited about this and doing your PhD right now, this is for you

Cartesia

@cartesia_ai

21 days

We're now hiring PhD research interns for spring/summer 2025 to work on architecture research and model training at Cartesia. You'll be part of a small team led by @_albertgu that's pushing the frontier of architecture research in AI. Apply here

2

21

118

2

53

Karan Goel

@krandiash

20 days

demo of our new localization feature, take one voice and localize to any language, accent or dialect technically shipping next week, couldn't resist showing it off since it's already live on our playground on our way to models that can control every aspect of voice perfectly

swyx @ DevDay!

@swyx

20 days

@goClueso Cartesia - @krandiash demoing accents/multilingual voice

1

32

1

4

51

Karan Goel

@krandiash

2 years

🚀 ChatGPT / GPT-4 for querying and asking questions on codebases Point to any GitHub repo, and get an index that is used to answer questions. Use --prompt-only mode if you can only access GPT-4 via ChatGPT to copy-paste. Built with @MeerkatML !

3

7

48

Karan Goel

@krandiash

3 months

nice start to the week closing two amazing engineers in two days 🏃🏃🏃

2

0

48

Karan Goel

@krandiash

19 days

this is why I built Sonic; no wisdom teeth and I can still talk on my way from the hospital doctors are amazing, but they could’ve thrown in a haircut for sure

6

1

49

Karan Goel

@krandiash

12 days

very proud of the work the team’s done so far in building cheap fast and high quality voice, we’ve gotten to openai quality in 3 months the next few updates that are coming are going to blow people’s minds

Cartesia

@cartesia_ai

12 days

Users come to us all the time with questions around how to evaluate the best voice generation APIs. To help, we put together a systematic comparison on the important features to look at when comparing Cartesia Sonic to ElevenLabs (link below) Another great resource is the

3

11

78

2

47

Karan Goel

@krandiash

4 years

We're bringing you the 2nd episode of the Stanford MLSys Seminar tomorrow. @matei_zaharia will talk about lessons from @databricks in building and deploying @MLflow . Tune in at 3pm PT Th at (and join our mailing list at )!

0

8

46

Karan Goel

@krandiash

4 years

If you told me I would be viral in 2020, I would 100% imagine being strapped into a ventilator in the ICU

1

0

43

Karan Goel

@krandiash

2 months

we worked with early partners to create telephony optimized versions of Sonic, and it's paying off we realized how voice sounds on a phone is very different from how voice sounds in an audiobook, and the optimizations we do make a huge difference to our telephony customers

Cartesia

@cartesia_ai

2 months

We're delivering high quality conversations with Sonic at the lowest latencies on voice ever seen with @Vapi_AI to their customers. Amazing to partner with them!

0

2

32

2

3

42

Karan Goel

@krandiash

3 years

A new tool in the Robustness Gym universe! This work is prompted by a basic lesson we’re learning in the RG project: quantitative metrics are fuzzy measures of performance, and need to be supported by interactive tools that support deeper inspection. Both are important!

Jesse Vig

@jesse_vig

3 years

Excited to announce SummVis, an interactive visualization 📊tool for analyzing summarization models 🤖, data 📰, and evaluation metrics📏. arXiv: code: w/ @iam_wkr @krandiash @nazneenrajani @SFResearch @StanfordAILab 1/N

6

45

197

1

15

42

Karan Goel

@krandiash

3 months

join the dark side it’s more fun here

anpaure

@anpaure

3 months

big tech is brain rotting

174

165

7K

0

41

Karan Goel

@krandiash

3 months

great summary of the talk I gave today at @aiDotEngineer by @kozerafilip building intelligent machines in the image of humans is a long and hard road, new ideas are going to get us there in the long run

Filip Kozera

@kozerafilip

3 months

Great talk from @krandiash from @cartesia_ai at @aiDotEngineer on how State Space Models can enable real-time multimodal intelligence. Let's dive in: 1. Real time on device intelligence will enable a multitude of different agents, doing things for you in the background, see

0

1

29

0

3

40

Karan Goel

@krandiash

6 years

I'm delighted that I'll be interning over the summer at @salesforce research! Looking forward to collaborating with @RichardSocher , @CaimingXiong and everyone else ^_^

0

4

38

Karan Goel

@krandiash

3 years

New preprint from us ( @_albertgu , @chrisdonahuey , @HazyResearch ), tweet 🧵 coming tomorrow (I’m not fast enough 😅)

Aran Komatsuzaki

@arankomatsuzaki

3 years

It's Raw! Audio Generation with State-Space Models Achieves SotA perf on autoregressive unconditional waveform generation. proj: repo: abs:

1

23

160

0

2

39

Karan Goel

@krandiash

3 months

as promised, we shipped 2 models and 3 new features today emotion control by API, timestamps on generations and no length limit anymore new models sound really good and we'll keep updating them, SSMs work! also working on wacky stuff with SSMs that I'm excited about 👾

Cartesia

@cartesia_ai

3 months

Huge Sonic release! 🇺🇸 New English model reduces breathiness & artifacts. 🌎 New multilingual model improves pacing, loudness & word error rate by upto 50%. 💨😡😳😁 Voice control API to precisely control speed and emotion. ⏰ Word timestamps on gen audio, use for captioning

4

10

86

3

1

39

Karan Goel

@krandiash

2 years

New blogpost on @StanfordCRFM : What will it take to put models like GPT-X into software and not have to worry about insane behavior and bugs? We discuss making foundation models a reliable software abstraction: new programming tools are going to be key!

1

8

39

Karan Goel

@krandiash

6 days

We're topping another third-party evals leaderboard with our Sonic model. Sonic is high quality, ultra fast and cheap for speech generation, and we're seeing amazing adoption along pretty much every use case and sector imaginable. And 3 new releases are coming.

Labelbox

@labelbox

6 days

Speech generation is a fascinating domain as it needs to be heard and felt in order to evaluate the true difference. We’re seeing a large variance in quality among the model providers. Generative AI companies like @cartesia_ai and @elevenlabsio put up impressive performance

1

0

5

1

5

41

Karan Goel

@krandiash

2 months

intern applications are open at Cartesia our interns get to work on projects that are actually important to us rather than on side quests apply at the link below!

Cartesia

@cartesia_ai

2 months

We're recruiting machine learning research interns for fall 2024, apply below by August 24th. Join us to build and ship cutting-edge multimodal models, and have fun along the way!

0

4

33

3

2

38

Karan Goel

@krandiash

2 months

🚢🚢🚢🚢🚢 and new products coming in Aug, can't wait to share the cool stuff the team is building

Cartesia

@cartesia_ai

2 months

More updates at Sonic! 📈Enhanced voice cloning to preserve speaker accents and tones even more 🗣️ Improved default voices on playground for loudness and clarity 🌎 New multilingual model reduces word error rate and improves prosody significantly 📞 Improved clarity and

4

10

70

1

0

37

Karan Goel

@krandiash

7 months

Someone pointed me to this fragment from Jensen's Wired article -- amazing to see the support around SSMs (and really cool that he's so technically plugged in)

0

5

35

Karan Goel

@krandiash

3 months

a lot of our users building voice agents asked for this, enjoy!

Cartesia

@cartesia_ai

3 months

We've shipped continuations 🐍, our most requested API feature. Sonic can now stream in text (e.g. LLM generations), and generate audio smoothly across chunks using the power of Sonic's state. This unlocks long transcript use cases, and real-time conversational voice agents!

5

14

105

3

0

34

Karan Goel

@krandiash

3 months

multilingual support is a big feature request for Sonic, we 🏃 very fast and shipped the first version in a few weeks, expect new updates here we're cooking some insanely cool models that are further out that will be a step change in speed, quality and capability

Cartesia

@cartesia_ai

3 months

Release day with 2 new models 🇺🇸 Sonic English Improved pacing, voice cloning and pronunciation, same 135ms latency 🌎 Sonic Multilingual 6 new languages (German, French, Spanish, Portuguese, Chinese, Japanese) with a new multilingual voice library And 🩺 HIPAA compliance

7

16

132

0

34

Karan Goel

@krandiash

4 years

Wow, this went randomly viral and seems to have struck a chord. In the spirit of self-promotion: check out our work on making ML models more robust. Video: Arxiv preprint: Very excited about the future of AI!

Model Patching: Closing the Subgroup Performance Gap with Data...

How can domain experts that identify problems in their classification models fix them? We introduce model patching, a framework for improving the performance...

www.youtube.com

2

33

Karan Goel

@krandiash

3 years

Come by our Model Patching poster at @iclr_conf today! We describe how data augmentation with a domain-translation model and combined with robust training can improve worst-case performance. Talk/Poster Link: Time: Today (Monday 5/3) 5-7pm PT [Spot C1]

0

4

33

Karan Goel

@krandiash

2 years

A very short blog post on 3 directions for data tools I’m personally excited about in the era of GPT-4. We’re working on these in @MeerkatML (stay tuned for something cool coming soon!)

The Future of Data Tools | Notion

How will data work change? I’m particularly interested in the work of knowledge workers: scientists, engineers, data analysts, lawyers, doctors—really anyone whose life is directly impacted by the...

krandi.notion.site

0

3

31

Karan Goel

@krandiash

3 years

We've crowdsourced a ton of contributions to so far! You can now get a broad overview of Data Centric AI there -- we've got discussion on weak supervision, self supervision, robustness, data augmentation, privacy, data selection, and more. Check it out!!

GitHub - HazyResearch/data-centric-ai: Resources for Data Centric AI

Resources for Data Centric AI. Contribute to HazyResearch/data-centric-ai development by creating an account on GitHub.

github.com

1

7

33

Karan Goel

@krandiash

2 years

I miss the pre-mid-2022 days when my Twitter was a daily digest of ML research preprints Now you can’t go 3 tweets without somebody trying to teach you a new incantation to yell into the magic box — it’s the AI equivalent of drugs and vegetables

1

2

33

Karan Goel

@krandiash

1 year

Excited to see the RedPajama dataset released: check out the @MeerkatML data exploration dashboard we put together in a collaboration with @togethercompute as part of this release 🚀 We’ll continue to update and add to that in the RedPajama repo!

Together AI

@togethercompute

1 year

Announcing RedPajama — a project to create leading, fully open-source large language models, beginning with the release of a 1.2 trillion token dataset that follows the LLaMA recipe, available today! More in 🧵 …

38

407

2K

0

9

30

Karan Goel

@krandiash

3 months

a nice post about the work that's warming up in ai new architectures will enable new problems to be solved, llms are the start

ashu garg

@ashugarg

3 months

I’ve been in the AI trenches since 2009, and LLMs are certainly a game-changer. But they also seem to be a warm-up act for the main event—the next cycle of AI innovation, coming in the next 12-18 months. Here are 3 areas we’re looking at to fuel this cycle, where founders can

46

131

781

1

3

29

Karan Goel

@krandiash

4 months

@AstleDsa SSMs generally crush on data derived from continuous signals -- we've observed this consistently across many applications and modalities (audio, video, EEG, EKG, other time series). Lots more to learn and improve here

3

0

29

Karan Goel

@krandiash

3 months

super cool demo using our super fast model

Guillermo Rauch

@rauchg

3 months

an oss voice assistant that pipelines state-of-the-art high-performance ai models: @groqinc whisper → llama3 → @cartesia_ai sonic

24

50

569

2

3

29

Karan Goel

@krandiash

3 months

really fun to hang out with @saranormous and @eladgil and shoot this episode of the @NoPriorsPod we cover a lot of ground across research, engineering and the future of ai systems and I preview some of our on device work with a demo

sarah guo // conviction

@saranormous

3 months

🔥 new @NoPriorsPod : @krandiash @_albertgu from @cartesia_ai : *state space models (SSMs) *their advantages, disadvantages *alternative architectures to transformers *making AI real-time (demo!) *sonic, audio applications *multimodality and research directions 🔗 in comment

3

13

64

0

4

28

Karan Goel

@krandiash

2 months

it’s been amazing to work with @jordan_dearsley & @nikhilro_ from day one at @cartesia_ai they’re amazing founders – experts in voice, move at breakneck speed and always put their customers first we’re working closely with @Vapi_AI to deliver exceptional voice agents to users

Cartesia

@cartesia_ai

2 months

@Vapi_AI switched to Cartesia as their default voice provider over 8 options after customers saw a 4x increase in call retention. Read how we built the best API for realtime conversation with this leading voice platform Try it at

0

7

39

1

2

28

Karan Goel

@krandiash

4 years

With all the demos flowing for GPT-3, I thought it would be fun to speculate about what this means for the future of user interfaces. I haven't blogged before, but I decided to take the plunge (it's short). GPT-3 & The Natural Language Programmer

GPT-3 and The Natural Language Programmer | Notion

TLDR: The only programming language you might need to know in the future is your native tongue.

krandi.notion.site

0

3

27

Karan Goel

@krandiash

2 years

ChatGPT is pretty cool. Braindump: It might make mistakes in reasoning and knowledge retrieval, this is not worth overindexing on in my opinion. This is certain to improve quickly, although it’s good to know what’s not working quite as well yet

1

4

26

Karan Goel

@krandiash

6 years

Delighted to announce that our paper (with @AIforHI ) on “Learning Procedural Abstractions and Evaluating Discrete Latent Temporal Structure” has been accepted to ICLR 2019!

1

27

Karan Goel

@krandiash

2 years

We’ve got new work out, appearing at NeurIPS this year! We extended S4 beyond sequences to handle images and videos. Our new S4ND layer is a drop in for ConvND in any architecture!

S4ND: Modeling Images and Videos as Multidimensional Signals Using...

Visual data such as images and videos are typically modeled as discretizations of inherently continuous, multidimensional signals. Existing continuous-signal models attempt to exploit this fact by...

arxiv.org

2

4

27

Karan Goel

@krandiash

2 years

Amidst a barrage of great work released at frenetic pace, it's easy to feel like there's nothing left for you to do (esp. in academia). I rarely worry about this now -- a trick I use is to imagine myself 3 years ago and then think about all the cool shit that has happened since.

1

25

Karan Goel

@krandiash

1 month

can't wait until we have always-on assistants that live on-device, understand language, audio and video and have multi-year memories -- basically a human

Cartesia

@cartesia_ai

1 month

Check out this fully on-device interview assistant running with our open source repo Edge and our new model Sonic On-Device 📲

3

4

60

1

25

Karan Goel

@krandiash

4 years

🎇Excited that our work on model patching has been accepted to ICLR 2021!

Karan Goel

@krandiash

4 years

Preprint alert! "Model Patching: Closing the Subgroup Performance Gap with Data Augmentation" is now on arXiv! 📑Paper: 🧑‍💻Code: 📹Video: ✍️Blog: Read on to learn more (1/9)

3

37

154

0

1

25

Karan Goel

@krandiash

2 years

The firehose of AI happenings has become so monstrously large I could scroll all day on Twitter and still be behind

1

25

Karan Goel

@krandiash

1 month

new work from one of our interns @avivbick on distillation into ssms stay tuned for more

Cartesia

@cartesia_ai

1 month

💡Research Spotlight: @avivbick Aviv, one of our summer interns, co-authored MOHAWK, a new way to distill quadratic Transformers into subquadratic architectures like SSMs. We’re proud of him for the groundbreaking research he conducted with Prof. @_albertgu our Chief Scientist

0

3

42

1

24

Karan Goel

@krandiash

1 month

we've opened up the private beta for Sonic on-device Sonic is the fastest cloud voice gen model (), and we've squeezed all of these capabilities into a model you can run locally fill out this form if you want to get early access

Cartesia On Device Waitlist

Turn data collection into an experience with Typeform. Create beautiful online forms, surveys, quizzes, and so much more. Try it for FREE.

bcpeugvkng2.typeform.com

Cartesia

@cartesia_ai

1 month

📲Sonic On-Device: The Sonic model you know and love, running on device in private beta. It’s the first ultra-realistic voice model of its kind to support real-time streaming on devices.

4

5

62

1

2

24

Karan Goel

@krandiash

25 days

this is such an insanely cool and fun use of voice AI

kwindla

@kwindla

25 days

An Homage To Metal Gear Solid a playable voice AI puzzle game <overheard in slack> me: I wrote some sample code to show how you switch out LLM context on the fly and why you might want to. @JonPTaylor : hold my beer ... </> Tech stack: - input speech processing @DeepgramAI

6

16

66

1

3

24

Karan Goel

@krandiash

4 months

@swyx @elevenlabsio @cartesia_ai @_albertgu Thanks! Lots to learn for us, @elevenlabsio is a company we all look up to We’re working hard to make our models even faster, there’s a lot of room left on the table — stay tuned

1

0

22

Karan Goel

@krandiash

4 months

my anecdotal experience is that Chinese researchers have insane velocity in adopting and iterating on new research, there are 25 follow ups to published work from the US in 3 months the US needs to level up fast, the EU is the EU

Arnaud Bertrand

@RnaudBertrand

4 months

So what proves that China has now become the world's foremost scientific power? Firstly, China has now overtaken both the US and entire EU in number of high-impact scientific papers produced each year, including in the Nature Index which is virtually impossible to game.

58

511

2K

2

24

Karan Goel

@krandiash

1 month

@reach_vb we've got more coming for our open-source friends haha

3

1

23

Karan Goel

@krandiash

4 years

Congratulations to @KabirGoel , who is headed to Cal for his undergrad!!

kabir 🧩

@KabirGoel

4 years

Committed to UC Berkeley over Duke. Hardest decision of my life thus far. Here’s to hoping I get out of this alive (and with all my limbs intact). Go Bears! 🐻

10

0

81

0

23

Karan Goel

@krandiash

3 years

Check out Mistral, our code base for training large LMs. We’ve also released multiple random seeds, 600+ checkpoints per run for GPT-2 Small and Medium

Siddharth Karamcheti

@siddkaramcheti

3 years

We're excited to open-source Mistral 🚀 - a codebase for accessible large-scale LM training, built as part of Stanford's CRFM (). We're releasing 10 GPT-2 Small & Medium models with different seeds & 600+ checkpoints per run! [1/4]

4

102

380

0

4

23

Karan Goel

@krandiash

1 month

@levelsio @reisr @elevenlabsio fastest latency in the world, higher quality, cheap and we run locally we're 3 months old, give it a whirl

1

0

22

Karan Goel

@krandiash

5 years

Super excited to get this grant with @HazyResearch and Sharon Li on new directions for robust machine learning systems. Shout out to @StephanZheng and @nazneenrajani for their support!

Salesforce AI Research

@SFResearch

5 years

We're thrilled to announce this year's @SFResearch Deep Learning Grant winners @ChenhaoTan @gregd_nlp @pulkitology Christopher Ré and Hung-yi Lee! 🎉👏 We're excited to work together to advance the state of AI. Read more about the winning proposals:

0

19

73

0

23

Karan Goel

@krandiash

18 days

tweets post anesthesia go crazy

Karan Goel

@krandiash

19 days

this is why I built Sonic; no wisdom teeth and I can still talk on my way from the hospital doctors are amazing, but they could’ve thrown in a haircut for sure

6

1

49

2

1

23

Karan Goel

@krandiash

4 months

@reach_vb We're cooking 🧑‍🍳

3

0

22

Karan Goel

@krandiash

3 years

Excited to see this report on foundation models go out today, where I co-authored the data section led by @laurel_orr1 : Huge credit to @percyliang and @RishiBommasani for orchestrating this and making sure each section hit a pretty stringent quality bar.

3

4

22

Karan Goel

@krandiash

5 years

My (17 yr old) brother just released his first product! It’s a Chrome extension that improves your exposure to news stories from other points of view. The design is great and it runs the latest and greatest NLP models for news recs. Download and review!

kabir 🧩

@KabirGoel

5 years

Unslant, my browser extension to surface ideologically contrasting takes on the political news you’re reading, is live on @ProductHunt ! First product release ever, can’t wait to see where this goes 😃

0

6

19

1

2

22

Karan Goel

@krandiash

2 years

this isn’t a business, but this might be a hobby

2

0

21

Karan Goel

@krandiash

1 month

art

0

22

Karan Goel

@krandiash

6 years

Come see our live demonstration tomorrow at #NeurIPS and see if you too can learn some Bach on our piano!

Stanford AI for Human Impact

@AIforHI

6 years

At 10:35 on Wed (Dec 5) @krandiash , Tong Mu, @turingmusician and Emma Brunskill will have a demo on “Automatic Curriculum Generation Applied to Teaching Novices a Short Bach Piano Segment” in Room 510 ABCD # D10

1

0

6

2

9

21

Karan Goel

@krandiash

4 months

Audio as a platform is going to be crazy I've talked to people who thought voice is on the way out - have you ever seen a human?

Jakub Neander (Zaiste)

@zaiste

4 months

Playing with @cartesia_ai and I’m super impressed! The voices feel natural and more human-like compared to anything I’ve seen before. Check out the demo! There are two interesting moments around 1:06 and at the end – not sure what happened there 🤪 Still measuring things but

3

2

30

7

0

21