all AI news
The Expressive Power of Transformers with Chain of Thought
March 21, 2024, 4:43 a.m. | William Merrill, Ashish Sabharwal
cs.LG updates on arXiv.org arxiv.org
Abstract: Recent theoretical work has identified surprisingly simple reasoning problems, such as checking if two nodes in a graph are connected or simulating finite-state machines, that are provably unsolvable by standard transformers that answer immediately after reading their input. However, in practice, transformers' reasoning can be improved by allowing them to use a "chain of thought" or "scratchpad", i.e., generate and condition on a sequence of intermediate tokens before answering. Motivated by this, we ask: Does …
abstract arxiv chain of thought cs.cc cs.cl cs.lg cs.lo graph however machines nodes power practice reading reasoning simple standard state thought transformers type work
More from arxiv.org / cs.LG updates on arXiv.org
Sliced Wasserstein with Random-Path Projecting Directions
2 days, 3 hours ago |
arxiv.org
Learning Extrinsic Dexterity with Parameterized Manipulation Primitives
2 days, 3 hours ago |
arxiv.org
The Un-Kidnappable Robot: Acoustic Localization of Sneaking People
2 days, 3 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Data Engineer
@ Lemon.io | Remote: Europe, LATAM, Canada, UK, Asia, Oceania
Artificial Intelligence – Bioinformatic Expert
@ University of Texas Medical Branch | Galveston, TX
Lead Developer (AI)
@ Cere Network | San Francisco, US
Research Engineer
@ Allora Labs | Remote
Ecosystem Manager
@ Allora Labs | Remote
Founding AI Engineer, Agents
@ Occam AI | New York