GitHub
LSF_MDia
观察池 · 暂无正式排名
该项目当前未满足“两个有效维度 + 两种数据源”的主榜门槛。
— 未排名
可观测采用度缺失 —
动量当前有效 · 2026-09-12 6
关注度当前有效 · 2026-09-12 0
信号可信度 依据当前有数据的独立评分维度数量计算。
中2/3 · 1 种数据源
项目介绍
Let LLMs invent, evolve, exchange, and route compact machine dialects for efficient reasoning.
Machine Dialectology (MDia) studies how heterogeneous LLM agents can create, inherit, exchange, and route compact symbolic languages for reasoning. Instead of forcing every model to express intermediate reasoning as long natural-language Chain-of-Thought, this repository explores Language Symbolism Frameworks (LSFs): reusable, machine-oriented symbolic protocols that compress reasoning into compact operators, schemas, constraints, and routing contracts.
Communication pressure + reuse + selection → emergent machine dialects.
Long natural-language rationales are often useful for humans, but they…
各数据源
424 Star
- Star 424
- Fork 15
- 提交 24
- 发布 0