Why isn’t LLM reasoning done in vector space instead of natural language?
·
0 reactions
·
0 comments
·
9 views
Why don’t LLMs use explicit vector-based reasoning instead of language-based chain-of-thought? What would happen if they did? Most LLM reasoning we see is expressed through language: step-by-step text, explanations, chain-of-thought style outputs, etc. But internally, models already operate on high-dimensional vectors. So my question is: Why don’t we have models that reason more explicitly in latent/vector space instead of producing intermediate reasoning in natural language? Would vector-based
Original article
LocalLlama
Anonymous · no account needed
Discussion
0 commentsMore from LocalLlama
Blog: AI evals are becoming the new compute bottleneck
LocalLlama·4
Qwen3.6-27B - Closed-loop SVG Images
LocalLlama·2
MiMo-V2.5-Pro - the actual best open-weights model
LocalLlama·4
Update on 2019 Mac Pro
LocalLlama·7
16x Spark Cluster (Build Update)
LocalLlama·7
daily ritual at this point…
LocalLlama·20