Lightnews — Scholar-powered news

Basis @basisresearch.bsky.social · Jul 18

Finally, Basis is hiring! Explore open roles here: www.basis.ai/join-us/#car...

Join Us

Careers at Basis

www.basis.ai

Basis @basisresearch.bsky.social · Jul 18

We’re also hosting a happy hour social for workshop participants and others interested in what we’re working on. If you’re at ICML and would like to join us *July 18th, 6-9pm*, RSVP here: lu.ma/gu2nea9q

Basis Social at ICML · Luma

Join us for an evening of conversation, networking, and some well-deserved drinks and bites! We're excited to bring together two ICML Workshops for this event:…

lu.ma

1

Basis @basisresearch.bsky.social · Jul 18

We’re presenting the benchmark and our latest experimental results *today* at two relevant ICML workshops we’re sponsoring: Programmatic Representations for Agent Learning (PRAL) and Assessing World Models.

1

Basis @basisresearch.bsky.social · Jul 18

Read more and interact with AutumnBench environments on our blog: www.basis.ai/blog/autumn-...

Autumn 2025: Benchmarking Human and AI Performance on World Model Discovery

Introduction When you first encounter a new device—be it a smartphone, kitchen appliance, or unfamiliar video game—you rapidly build an intuitive mental model of how it works. Within minutes, you can ...

www.basis.ai

1

Basis @basisresearch.bsky.social · Jul 18

We’re proud to announce the launch of AutumnBench, an open-source benchmark developed on our Autumn platform. This benchmark, led by our MARA team, provides a novel platform for evaluating world modeling and causal reasoning in both human and artificial intelligence.

1 1

Basis @basisresearch.bsky.social · Jul 18

We’re presenting the benchmark and our latest experimental results *tomorrow* at two relevant ICML workshops we’re sponsoring: Programmatic Representations for Agent Learning (PRAL) and Assessing World Models.

Basis @basisresearch.bsky.social · Jul 18

AutumnBench is open-source and available now: autumn.basis.ai

Autumn

autumn.basis.ai

1

Basis @basisresearch.bsky.social · Jul 18

Read more and interact with AutumnBench environments on our blog: www.basis.ai/blog/autumn-...

Autumn 2025: Benchmarking Human and AI Performance on World Model Discovery

Introduction When you first encounter a new device—be it a smartphone, kitchen appliance, or unfamiliar video game—you rapidly build an intuitive mental model of how it works. Within minutes, you can ...

www.basis.ai

1

Basis @basisresearch.bsky.social · Mar 20

We are hiring for roles within this project team and others. Learn more and apply: www.basis.ai/join-us/#car... (3/3)

Basis @basisresearch.bsky.social · Mar 20

This work will apply Basis core technologies in probabilistic simulation and program synthesis to further ARIA's and Basis' mutual goal: to rapidly accelerate robot hardware development. (2/3)

1

Basis @basisresearch.bsky.social · Mar 20

Basis is proud to be an R&D creator in @aria-research.bsky.social's Robot Dexterity programme, led by Programme Director @jennycaread.bsky.social. Our team is developing a rational automated design agent that can perform end-to-end co-design of robot morphologies with their controllers. (1/3)

Advanced Research + Invention Agency (ARIA) @aria-research.bsky.social · Mar 20

From developing a next-gen electronic skin for robots, to crafting novel tech for muscle-like actuation, introducing the next seven teams we’re funding in Robot Dexterity. (1/2)

1

Basis @basisresearch.bsky.social · Dec 2

New whitepaper with contributions from Basis scientists

Patrick Mineault @patrickmineault.bsky.social · Dec 2

Excited to release what we’ve been working on at Amaranth Foundation, our latest whitepaper, NeuroAI for AI safety! A detailed, ambitious roadmap for how neuroscience research can help build safer AI systems while accelerating both virtual neuroscience and neurotech. 1/N

Different paths toward safe AI at different Marr's levels

6