AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore65

ARCHER: From Rationale Imitation to Executable Reasoning in Small Language Models

Summary

This work moves small language models from imitating reasoning traces to executable reasoning, making them more reliable on math tasks.

Why this matters

Covers model and deploy · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items