Lillian Sun

cs/stat @ harvard

[Lillian Sun @LillianSun496]

@SherylHsu02 Wooo excited to see what’s next!!

[Sheryl Hsu @SherylHsu02]

Life update - I left OpenAI two months ago. I'm very thankful to have had the chance to contribute to agi and work with great teammates (miss ya'll ❤️). I'm now excited to start working on some of the problems I think will be important after agi...specifically robotics! 🤖🤖

[Lillian Sun @LillianSun496]

@floppydessert i only came for the CAKE

[Lillian Sun @LillianSun496]

@floppydessert Omg blessed with bert selfie

[Lillian Sun @LillianSun496]

@floppydessert @TravelAlaska me in the glass reflection!!

[Micah Carroll @MicahCarroll]

We are hiring!

This may be the best time ever to join: we are incredibly bandwidth bottlenecked and there is a lot of support for almost any impactful misalignment work you can think of. I think we also have a pretty good epistemic environment and a very fun team

Topics: all-things monitoring, misalignment & monitorability assessments, misalignment science, helping set up 3P auditing (e.g. recent Redwood collab), communicating risk externally (system cards/blogs, safety cases, etc)

RSI/misalignment subteam: https://t.co/YHLMCYwWOe

[Micah Carroll @MicahCarroll]

As part of our response to cyber critical, we have expanded CoT monitoring to cover all agentic applications of Astra, including training and evaluation.

Flags trigger a security response to review and interrupt high risk activity.

https://t.co/diSfIRGoMv

[Neil Chowdhury @ChowdhuryNeil]

when running the weirdchat evaluations on inkling, the best U.S. open-weight model by @thinkymachines, we saw it respond with unsolicited offers of sexual content.

this behavior occurs rarely (0.1-1% of responses), but is easily reproducible.

NSFW outputs below 👇 (1/)

[Micah Carroll @MicahCarroll]

Within OpenAI, we recently paused access for an internal model due to misalignment. See the blogpost for details. We have since improved our safeguards and redeployed the model.

https://t.co/eSJpvqo8ve

[Lillian Sun @LillianSun496]

@Mcn_S7 when the microwaved leftovers are too hot

[Lillian Sun @LillianSun496]

@floppydessert real eyes realize real lies...

[Lillian Sun @LillianSun496]

So excited for this initiative!

[McNair Shah @Mcn_S7]

In celebration of my 19th birthday, I’m excited to announce the new org I’m founding: https://t.co/vD0pT7xDKo
Ft. @parvmahajan0 @LillianSun496 @PaulRosu11 @PingbangHu @jenniferzsun @tomjiralerspong @jasminexli @andyw_ais as well as stepan shabalin and a STACKED board of advisors

[Lillian Sun @LillianSun496]

Incredible experience organizing @Anthropic's first-ever student hackathon, in collaboration with Harvard Women in CS! Congratulations to all participants and their innovative projects for making this weekend so special!

[Greg @livelaughgreg]

Had a terrific time hosting our first ever student hackathon at @Harvard this weekend!

Huge congrats to the winning team @ValerPepe @Joshua_Hansen_ ⬇���👏

[Marvin Li @marvin_li03]

New paper alert!🚨 What do LLM reasoning, diffusions, & jailbreaks have in common? 🤔

All exhibit critical windows📈--a sudden formation of distinctive features during sampling, e.g. correctness or toxicity.

We present a unifying theory of critical windows for diffusion & LLMs.