Joseph Lai 🍱

Co-Founder & CEO at Bentocart (YC W19)

[Joseph Lai 🍱 @jtlai]

"You're absolutely right!"

OK, I feel this is too assuring 😅 especially when hearing it so many times per day.

[Joseph Lai 🍱 @jtlai]

@borisjabes @census @fivetran @frasergeorgew Wow congrats Boris and Census team!

[Cameron Jones @camrobjones]

New preprint: we evaluated LLMs in a 3-party Turing test (participants speak to a human & AI simultaneously and decide which is which).

GPT-4.5 (when prompted to adopt a humanlike persona) was judged to be the human 73% of the time, suggesting it passes the Turing test (🧵)

[Joseph Lai 🍱 @jtlai]

This should have been just a line item under the payment or invoice object as "Invoice Fee". But they deliberately choose to hide it. Very disappointed.

[Joseph Lai 🍱 @jtlai]

PSA: If you use Stripe Invocing, they've been taking 0.4% of your invoice total silently.

These fees don't show up in the invoice itself, nor the corresponding payment. It's hidden in an obscure "balance report". This is NOT transparent at all.

[Ivan Leo @ivanleomk]

Processing PDF files to chat with in 2023 was an entire startup idea. Today with Gemini's native multimodality, all it takes is a simple file upload to get started.

You can now process audio, videos, PDFs and even images with a simple upload using gemini instead of spending valuable engineering hours on a pre-processing pipeline. The 2 million token context also makes it easy to stuff a gazillion things inside it.

Instructor makes this easy with its automatic validation and retries of the response, turning messy and sometimes unreliable llm outputs into clean, validated Pydantic objects.

Building a prototype shouldn't have to feel like rocket science and starting with structured outputs will save you hours down the line - whether it's for testing, future fine-tuning or even manual annotation.

Check out our latest article that walks you through how to do just that

https://t.co/oR08grdJ35

[Rohan Paul @rohanpaul_ai]

Local models now protect your privacy while still accessing powerful LLM capabilities

Chain small and large LLMs to get best performance while keeping data private

🔍 Original Problem:

Users share sensitive personal information with proprietary LLMs during inference, raising privacy concerns. While local open-source models help with privacy, they perform worse than proprietary models.

-----

🛠️ Solution in this Paper:

• PAPILLON: A multi-stage pipeline where local models act as privacy-conscious proxies
• Uses DSPy prompt optimization to find optimal prompts for privacy preservation
• Two key components:
- Prompt Creator: Generates privacy-preserving prompts
- Information Aggregator: Combines responses while protecting PII
• Created PUPA benchmark with 901 real-world user-LLM interactions containing PII

-----

💡 Key Insights:

• Simple redaction significantly lowers LLM response quality
• Privacy-conscious delegation can balance privacy and performance
• Smaller local models can effectively leverage larger models while protecting privacy
• Prompt optimization improves both quality and privacy metrics

-----

📊 Results:

• Maintains 85.5% response quality compared to proprietary models
• Restricts privacy leakage to only 7.5%
• Outperforms simple redaction approaches
• Shows consistent improvement across different model sizes

[Joseph Lai 🍱 @jtlai]

🤙

[Lisa Dunlap @lisabdunlap]

🧵We love measuring accuracy, but what about the vibes?

Intro VibeCheck—a system that discovers and measures qualitative differences in LLMs. VibeCheck shows Llama is friendlier while GPT-4 focuses on ethics; these vibes can even predict model identity and user preference.

https://t.co/PpeJEcc7gj

[Zhe Gan @zhegan4]

💡Imagine a multimodal LLM that masters universal UI understanding across platforms?

Here it is, 🎁 we upgrade Ferret-UI to Ferret-UI 2, a generalist model for grounded UI understanding across iPhone, Android, iPad , Webpage, and AppleTV. Check the image below for visual examples.

And check our paper for details on how we achieve this: https://t.co/bshfQrIAuP.

Led by our awesome intern Zhangheng, together with Keen, @HaotianZhang4AI @yinfeiy and other great collaborators.

[Maziyar PANAHI @MaziyarPanahi]

Microsoft just dropped OmniParser model on ⁦@huggingface⁩, so casually! 😂

“OmniParser is a general screen parsing tool, which interprets/converts UI screenshot to structured format, to improve existing LLM based UI agent.” 🔥 https://t.co/h9nzhyUUQB

[Ruohong Zhang @RuohongZhang]

[p1] Improve Visual Language Model Chain-of-thought Reasoning

paper link: https://t.co/eUnlisUsv5

project page (to be updated upon approval on release): https://t.co/LpAYt6k8yQ

Content:
1. We distill 193K CoT data
2. Train with SFT
3. DPO to futher improve performance

[merve @mervenoyann]

I'm bullish on this foundation OCR model called GOT 📝 @eccvconf

This model can transcribe anything and it's Apache-2.0!

Keep reading to learn more 🧶

[Joseph Lai 🍱 @jtlai]

Creative and surprisingly it works

[Joseph Lai 🍱 @jtlai]

@tomkit08 Need one for property tax assessment appeal too!

[Nat Friedman @natfriedman]

"People hire a janitor service to clean their office. They don't hire a generic labor service, even though it's basically the same thing." – advice for AI startups.

[Joseph Lai 🍱 @jtlai]

@jeffiel 🧡 The Onion & ty @jeffiel for keeping them around!

[Joseph Lai 🍱 @jtlai]

A few of our domains started getting moved and it's a shit show. Squarespace's DNS update is slow & unreliable. Good luck trying to fix anything they mess up. (And meanwhile your whole service is down to your customers due to DNS issues)

[Joseph Lai 🍱 @jtlai]

PSA: If you have any domains on Google Domain, transfer them out NOW before the Squarespace move causes serious disruption to your service.

[Joseph Lai 🍱 @jtlai]

@chptung same