PhysicsLM: Physics as Token Prediction
PhysicsLM has been developed, utilizing autoregressive language modeling to predict 2D rigid body dynamics. It unifies physical modeling and language models through the concept of next-token…
Read the full articleYou might also wanna read
Recursive Language Models: A New Approach for Processing Extremely Long Prompts Beyond Standard Context Windows
We study allowing large language models (LLMs) to process arbitrarily long prompts through the lens of inference-time scaling. We propose Re
arxiv.org·6mo ago
Zebra-Llama: Efficient Hybrid Language Models Combining SSMs and Attention Layers
With the growing demand for deploying large language models (LLMs) across diverse applications, improving their inference efficiency is cruc
arxiv.org·7mo agoAccelerating Large-Scale LLM Inference on AMD Instinct MI350X/MI355X with Eagle3 and AMD Quark
Large language model (LLM) inference is increasingly constrained by autoregressive decoding. Even when prefill is highly optimized, the deco
Efficient Vision Encoding for Vision Language Models
Vision Language Models (VLMs) enable visual understanding alongside textual inputs. They are typically built by passing visual tokens from a
machinelearning.apple.com·11mo ago
Physics-Informed AI, Part III: From Post-Hoc Checker to Differentiable Physics Head
Part II checked physics after generation. Part III puts the residual inside the training loop with a language-conditioned numerical head. In
Multi-Stream LLMs: A Parallel Architecture to Overcome Single-Stream Bottlenecks in Language Models
The continued improvements in language model capability have unlocked their widespread use as drivers of autonomous agents, for example in c
arxiv.org·1mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.