We’re introducing ToyBlocks, an open simulated tabletop for controlled experiments in robot learning.
It enables robotics researchers to experimentally control training data & models, run targeted evals, & trace failures to their causes. 🧵
github.com/allenai/ToyBlocks
Open benchmarks help move AI forward. 📄
@UnsiloedHQ reports scoring 88.0 on our olmOCR-Bench, ahead of GPT-5.5 at 84.6.
Great to see our open benchmark helping the field measure progress on reading complex documents!
Explore the leaderboard: huggingface.co/datasets/alle…
We’ve raised a $7.5M seed round led by Felicis so AI can read documents as easily as it reads databases.
Finance, law and healthcare run on documents, and not databases like any other industry.
Soon AI agents will read and write more documents than humans, and they’ll need to query them the way software queries tables.
We’re building the computable layer that makes this possible.
To do this at scale, we've brought together ML researchers and engineers from Google DeepMind, Cognition, and quant trading firms.
We use computer vision and dual-stream models to turn complex multimodal documents into formats LLMs and AI agents can understand and reason over.
If your team wants to turn messy documents into reliable, AI-ready data, book a demo:
unsiloed.ai/book-demo
Geospatial AI is moving fast, and putting it to work for the public good takes more than any one group can do alone.
Ai2 and @TaylorGeoAI are teaming up on open benchmarks, models, and data that practitioners can build on. 🧵
allenai.org/blog/taylor-geos…
With the community, we’re also helping shape how geospatial AI is evaluated through Benchmarks of The World, so models are compared consistently and reliably.
The partnership plans to release its datasets, benchmarks, and model weights free on @source_coop under open licenses.
1
1
541
We want input from the people who use these benchmarks. If you build, evaluate, or depend on geospatial foundation models, tell us which ones every model should be tested against.
Read the announcement and share your input → allenai.org/blog/taylor-geos…
456
Asking how to thicken frosting sliding off warm cupcakes? 🧁 Let them cool first!
That’s the XY problem: seeking a fix that misses the real issue. XYBench tests whether LLMs spot & address it.
🏆 It won the Best Resource Award at EMNLP 2026! 🧵
github.com/allenai/XYBench
We built XYBench to test whether prompting or training changes lead to more useful advice from LLMs.
We hope it helps researchers build assistants that guide users toward helpful solutions, even when they don’t know the right questions to ask.
📝 Blog: allenai.org/blog/xybench
NVIDIA used our MolmoSpaces benchmark to evaluate GR00T 2, alongside RoboCasa365, RoboLab, & RoboDojo. Thanks @NVIDIARobotics!
MolmoSpaces is open, so any team can test their robot policies on the same simulated homes, objects, & tasks:
molmospaces.allen.ai/leaderb…
A finer touch for humanoid robots. 🤖
At #CoRL2026, NVIDIA is introducing GR00T 2, an open world action model for humanoids, alongside new open industrial benchmarks from @RLWRLD_ai and @UCBerkeley built on Isaac Lab-Arena.
Explore how NVIDIA’s open world action models, scalable simulation tools and agentic workflows help researchers train, test and refine robot skills, from connecting cables to assembling small components.
Discover more ➡️ nvda.ws/4ixvdZV
Ai2 retweeted
Ai2 / AI for Science is hiring Summer 2027 research interns! Topics:
- Tools for ideation and hypotheses development
- Long-horizon human-AI interaction
- Generative user interfaces
We encourage publishing and public deployments. Apply below by Dec10 for first review cycle 👇
Ai2 retweeted
World models are becoming simulators for evaluating and training robot policies. But how do we compare them?
They output latents, video or simulator states, and image quality says little about task success. In ToyBlocks we compare what matters: interactions and outcomes.
Introducing “Task-Relevant Fidelity in Robotic World Models,” our work on ToyBlocks, to be presented at the LEAP and TAG/G2A workshops at #CoRL2026.🧵
We’re introducing ToyBlocks, an open simulated tabletop for controlled experiments in robot learning.
It enables robotics researchers to experimentally control training data & models, run targeted evals, & trace failures to their causes. 🧵
github.com/allenai/ToyBlocks
Ai2 retweeted
We're hiring interns! :)
Please apply by November 15 (this Sunday) to be considered in the first round of reviews.
You can find more details about our research focus areas and how to apply on the application page!
Replying to @allen_ai
& we’re accepting intern applications in:
• AI for Science
job-boards.greenhouse.io/the…
• Open Ecosystem
job-boards.greenhouse.io/the…
• Robotics
job-boards.greenhouse.io/the…
We’re hiring across Ai2 👋
We’re growing our AI for science work:
• Research Scientist, Asta
job-boards.greenhouse.io/the…
• Young Investigator, Open Language Models for Biology
job-boards.greenhouse.io/the… 🧵
& we’re accepting intern applications in:
• AI for Science
job-boards.greenhouse.io/the…
• Open Ecosystem
job-boards.greenhouse.io/the…
• Robotics
job-boards.greenhouse.io/the…
Explore the roles & apply—or share with someone who might be a good fit.
→ See all openings: allenai.org/careers
6
14,250










