季广达 Guangda Ji

PhD Student in Compute Graphics at Simon Fraser University

[Francisco Gómez de Tejada | arquitecto @gomezdetejada]

ArtSpace, instalación de arte público inmersiva en Pekín (China).

Diseñada por Auspicious Art como una cubierta cinética de gran escala suspendida sobre una zona peatonal.

Compuesta por miles de cintas ligeras y elementos de malla, la estructura crea una superficie similar a una nube que varía según la luz solar, el viento y el punto de vista del observador.

[Keenan Crane @keenanisalive]

For 3D learning, the “Bitter Lesson” doesn’t properly respect Kolmogorov complexity.

The colloquial version of the bitter lesson is that, in a nutshell, just scaling up data and compute always beats problem-specific modeling (note that this is not actually what Sutton said; more on that later).

But look at the current situation with LLMs and agents generating 3D content: the really impressive examples depend on configuring an API, language, or software package (Three.js, OpenSCAD, Blender, etc.). Each representation impacts the description size for a given object—and for the classes of object that can be faithfully represented via one of these procedural descriptions, the encoding is vastly more efficient than an explicit description (points, splats, polygons, etc.) or even an implicit description (SDF, NeRF, etc.).

Moreover, the smaller you can make the procedural description, the better the LLM works: it has to predict fewer tokens to describe the same object, because each token does a better job of compressing geometric information. E.g., the words “unit sphere” are far more compact than a list of points or a collection of voxels.

So, in the procedural setting, building models for predicting shape becomes an exercise in bounding Kolmogorov complexity: what’s the smallest “program” I can use to describe a given object? (Or more precisely: which language minimizes description length *and* the entropy of the description, for a model trained on that language.)

That’s why current tools lean on carefully crafted abstractions (again, Blender, OpenSCAD, Three.js, etc.), for which many examples are available. Astra doesn’t run off and build you a giant neural network, hoping that the “right representation” will somehow be learned deep within the network weights.

None of this, by the way, contradicts what Rich Sutton actually said: the input/output encoding matters, even if the rest is just scaling up search and learning. And the reason we see examples in Three.js and not a more esoteric language is that there is a large corpus of examples to learn from. But the idea that the network will just “learn the best representation” by adjusting its weights is a misguided fiction.

Instead, it really seems like we are entering an era of “code as geometry” counterposed to WorldLabs’ “geometry as code”, where design of domain-specific languages is a key task. (And of course, search and learning will inevitably play a role in this design…)

Computer Graphics also plays a central role, because it’s the mechanism for decompressing the procedural description into an explicit representation (building a cylinder, tessellating a curve, etc.). That’s the tool calling part of the agentic process. And the better these graphics algorithms get, the more effective the language becomes. People are worried about the death of Computer Vision—but it seems Graphics is coming back from the grave. 🧟‍♂️

(And, might I add, that differential geometry is a very expressive language that few are taking advantage of in this setting. 😉)

[季广达 Guangda Ji @GuangdaJi]

For our SymTRELLIS sigasia26 paper, we also provide a cool demo on HuggingFace (https://t.co/Lo28PcaBVl)!

You can create your own symmetrized 3D mesh with this demo. We also recommend that you run it locally!

The following video is an instruction on how to play with it. The video is accelerated.

Checkout paper and code at https://t.co/Nw7yxTJAGW!

[季广达 Guangda Ji @GuangdaJi]

Thanks for @mingrui_zhao1 for this amazing video! 🤩

[季广达 Guangda Ji @GuangdaJi]

Symmetry! Symmetry! Symmetry!

Meet SymTRELLIS, symmetry-enforced 3D asset generation.

Rotational, reflectional or polyhedral, have them under your control without retraining.

SIGGRAPH ASIA 2026

Paper, code & demo available at: https://t.co/82Bl9gibwI

[Jianglong Ye @jianglong_ye]

Evolving dexterity with GPT-6 Astra 🖐️

Been trying Astra recently. Its zero-shot dexterous manipulation is already quite surprising. More interesting is seeing it learn and improve through simulation training, from pen spinning and Rubik's Cube to hammer use.

The real goal would be to evolve this dexterity in the real world.

https://t.co/GZjQTsklYc

[Mingrui Zhao @mingrui_zhao1]

3D models can look right and still fall apart.

Introducing Functionalization: graph completion adds missing structure and connectors to rectify motion and make models work.

SIGGRAPH Asia 2026
Paper, code, data available at: https://t.co/P5PYq6HFzU

[Qineng Wang @qineng_wang]

GPT-6 Astra can solve some seriously tricky spatial constraint tasks: from unlocking interlocked parts to threading a rope through three rings.

Try it yourself now 👇
https://t.co/him6NbTI2a

[Hao (Richard) Zhang @richardzhangsfu]

I will be giving two workshop talks at #ECCV2026 (stay tuned) and opening a third on Functionality, Articulation, and Interaction (https://t.co/JEmc8DsH6z). Please also check out our oral paper on PADFormer (https://t.co/zlzpBEZMyw) on Friday.

[Lucas Martinic @lucas_martinic]

14/100 I had to jump into this train, porting old games into newer platform is becoming easier and cheaper than ever with AI, would love to see @Nintendo or @PlayStation doing original, modern remakes!

PD: I own an original copy of the N64 game.

[鱼总聊AI @AI_Jasonyu]

行了，也确实是第一次为了一个男人熬夜、买token、做片。

五分钟，花了我不少钱，大家别吃文字版的瓜了，视频给你们准备好了～

孙割，你这新概念作文一等奖真不是盖的，你这故事后面拍电影必火，一定要找我合作🤝

[炼金叔叔 @AirdropAlchemis]

《磨指甲》

哪位天才做的？

[H.E. Justin Sun 👨‍🚀 🌞 @justinsuntron]

https://t.co/nLMTwWr67Z

[JY @dogewlfi]

找到了 牛来 电影的完整版本 大家不用担心被下架啦。

[Physics-astronomy @Physicsastronmy]

THIS Is What a Solar Eclipse Looks Like From Space 🌑

[Massimo @Rainmaker1973]

If you missed the solar eclipse, this is what it looked like:

[MERICA MEMED @Mericamemed]

Holy ai!

[Alexander Panfilov @kotekjedi_ml]

We can finally talk about it:

We found a way to extract hidden reasoning of frontier models using a vulnerability in the APIs of every frontier AI company.

We verified that our reasoning token count matches billed API thinking tokens 1:1 for most of the prompts we queried.

[The End Of All Our Exploring @RogerMelian]

Oh, no, ahora hay uno mejor 😭😭😭😭😭😭