Bin Ren

Exoplanet @ Xiamen. Private pilot. Previous @ObsCotedAzur ←@Caltech ←@JohnsHopkins / @STScI ← @xmuchina.

[Bin Ren @renbbin]

I do have postdoc openings in planet formation/tech/algos.

[NASA Solar System @NASASolarSystem]

New Horizons is healthy and cruising through the distant Kuiper Belt after the longest hibernation period of its mission yet. A recent beacon confirmed the spacecraft is operating normally more than 5.7 billion miles from Earth. 🚀🌌 https://t.co/jQhxf1Rcvr

[人民日報 People's Daily @PDChinese]

【長征十號乙首飛成功，中國運載火箭首次實現可控回收】2026年7月10日，長征十號乙運載火箭在海南商業航天發射場發射升空，將衛星順利送入預定軌道。

火箭一二級分離後，一子級垂直返回，在海上回收平台通過網繫捕獲方式回收，發射及一子級回收任務取得圓滿成功。

此次任務是中國首次成功實施運載火箭一子級可控回收，標誌著中國重複使用火箭技術取得重大突破。

[Bin Ren @renbbin]

@NASAUniverse If it works, the next step should be Hubble!

[Alex Veremeyenko @alex_verem]

This paper from Harvard and MIT quietly answers the most important AI question nobody benchmarks properly:

Can LLMs actually discover science, or are they just good at talking about it?

The paper is called “Evaluating Large Language Models in Scientific Discovery”, and instead of asking models trivia questions, it tests something much harder:

Can models form hypotheses, design experiments, interpret results, and update beliefs like real scientists?

Here’s what the authors did differently 👇

• They evaluate LLMs across the full discovery loop hypothesis → experiment → observation → revision
• Tasks span biology, chemistry, and physics, not toy puzzles
• Models must work with incomplete data, noisy results, and false leads
• Success is measured by scientific progress, not fluency or confidence

What they found is sobering.

LLMs are decent at suggesting hypotheses, but brittle at everything that follows.

✓ They overfit to surface patterns
✓ They struggle to abandon bad hypotheses even when evidence contradicts them
✓ They confuse correlation for causation
✓ They hallucinate explanations when experiments fail
✓ They optimize for plausibility, not truth

Most striking result:

`High benchmark scores do not correlate with scientific discovery ability.`

Some top models that dominate standard reasoning tests completely fail when forced to run iterative experiments and update theories.

Why this matters:

Real science is not one-shot reasoning.

It’s feedback, failure, revision, and restraint.

LLMs today:

• Talk like scientists
• Write like scientists
• But don’t think like scientists yet

The paper’s core takeaway:

Scientific intelligence is not language intelligence.

It requires memory, hypothesis tracking, causal reasoning, and the ability to say “I was wrong.”

Until models can reliably do that, claims about “AI scientists” are mostly premature.

This paper doesn’t hype AI. It defines the gap we still need to close.

And that’s exactly why it’s important.

[Bin Ren @renbbin]

@FlySJC let goo !!!

[Phil Armitage @philip_armitage]

With Ralph Pudritz, I've been editing the section of the 2nd edition of the Handbook of Exoplanets focused on "Formation and Evolution of Planets and Planetary Systems". This thread collects together the contributions currently available on astro-ph: (1/5)

[Bin Ren @renbbin]

@YinhaoW @lufthansa You can buy stuff online (if they deliver...)!

[Bin Ren @renbbin]

@GregHerczeg @cvspharmacy wtf 零元购？！！！

[Bin Ren @renbbin]

One of my proudest research works was with you, thank you and au revoir!

[ESA Gaia @ESAGaia]

So long and thanks for all the fish! I just observed my last star this morning... #GaiaDPAC and @esa Gaia teams will do their best to make some great data releases from all the data I gathered! #GaiaDR4 in 2026, #GaiaDR5 around the end of the decade. Keep posted for our news items today at 10:00 CET.

[Empire Of Lies @berningman16]

This image is so 2025.

[Zeke Hausfather @hausfath]

Great (and scary) visualization of 2024 daily temperatures compared to prior years by the BBC today. Evocative of the iconic Joy Division album cover from 1979: https://t.co/o2zLigFV7h

[Caltech Alumni Association @caltechalumni]

In response to this week's devastating Eaton Fire, we have established the Caltech and JPL Together Relief Fund to support our affected colleagues. Please consider making a contribution to this special fund. Thank you from all of us at Caltech. https://t.co/z5GxU2w0RF

[lee feinberg @leefeinberg1]

https://t.co/RIDEy8eQGe
Please help our many friends and colleagues recover from this terrible situation.

[Bin Ren @renbbin]

@GregHerczeg So that we can check if we had the same driver! That’s what made me want to get back my Maryland drivers license, then convert it to a European one to enjoy die Luft!

[Bin Ren @renbbin]

We can use coronagraph to image quasar surroundings, and explore the new frontiers on extreme physics! Thank you @NASAHubble!

[Hubble @NASAHubble]

Hubble took the closest look yet at the first known quasar.

Called 3C 273, this quasar is a brightly glowing galactic center powered by a black hole consuming material.

This helped astronomers open up a new gateway into better understanding quasars: https://t.co/16gBOXtWU5

[Misha Teplitskiy | Science of Science @MishaTeplitskiy]

People usually think replication attempts in science are rare. Journals don't publish replications, so scientists don't do them.

In reality there are countless replication attempts (and failures), it's just PhD students assume they did something wrong
https://t.co/cxGGexVOw2

[Bin Ren @renbbin]

@EricLagadec @EditionsduSeuil Le livre est tellement génial, que je savais pas que j'étais l'un des premières à avoir la chance de le voir avant sa sortie!