Pith. sign in

Llama-nemotron: Efficient reasoning models, 2025

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.CL 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

Thinkless: LLM Learns When to Think

cs.CL · 2025-05-19 · conditional · novelty 7.0

A 1.5B model learns to decide when to use short versus long reasoning via control tokens and a decoupled GRPO objective, reducing thinking-mode usage by 50-90% on math benchmarks with minor accuracy loss.

citing papers explorer

Showing 1 of 1 citing paper.

  • Thinkless: LLM Learns When to Think cs.CL · 2025-05-19 · conditional · none · ref 4

    A 1.5B model learns to decide when to use short versus long reasoning via control tokens and a decoupled GRPO objective, reducing thinking-mode usage by 50-90% on math benchmarks with minor accuracy loss.