Curious about how AI is starting to do science on its own? Come join us for this AI4Science talk: Data-driven Discovery at Ai2, Open-ended Scientific Discovery via Bayesian Surprise. At Ai2, this line of work has been building toward the idea of using AI to extract genuinely
@askalphaxiv
-
LeCun’s Temporal Straightening for Latent Planning Paper
By
–
Yann LeCun is pumping out papers recently “Temporal Straightening for Latent Planning” This paper shows that by straightening latent trajectories in a world model, Euclidean distance starts to reflect true reachable progress, so it's closer to geodesic/minimum-step distance.
-
Dense Task Experts in Pretrained Neural Networks Thickets
By
–
RL is no longer needed? "Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights" This paper argues that large pretrained models don’t sit at a single optimal set of weights but inside a dense “thicket” of nearby task-specific experts. So once pretraining is
-
Neural Cellular Automata Pre-Training for Language Models
By
–
Pre-Pre-Training on Neural Cellular Automata? This paper argues that LLMs may not need human text to learn core thinking machinery. So by first training on synthetic neural cellular automata, the model learns transferable rule-and-pattern inference that speeds up and improves
-
IndexCache Accelerates Sparse Attention via Cross-Layer Index Reuse
By
–
A really cool paper from @Zai_org "IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse" This paper shows that in sparse-attention LLMs, nearby layers usually pick almost the same important tokens. So you can cache and reuse those token indices instead of
-
Rubric-as-Reward Approach Advances Image Captioning Models
By
–
Really cool rubric-as-reward work for image captioning!
-
OpenClaw-RL: Train AI Agents Through Live Feedback
By
–
"OpenClaw-RL: Train Any Agent Simply by Talking" OpenClaw-RL’s big idea is that every time an AI agent gets a reply, error, test result, or tool output, it’s already receiving free feedback. So instead of wasting those signals, this paper turns everyday use into live
-
SentientAGI Partnership Launch for Arena Innovation
By
–
Excited to be partnering up with @SentientAGI for the Arena! We'll also be introducing some of the newest things we're working on
-
Dynamic Chunking Improves Diffusion Transformer Efficiency
By
–
“Dynamic Chunking Diffusion Transformer” Instead of splitting every image into the same grid of patches, this paper lets the model use bigger chunks for simple areas and finer chunks for detailed areas. It also changes the token budget over the diffusion process, where fewer