Self-aligned reward uses relative perplexity differences to encourage concise, query-specific reasoning in LLMs, yielding 4% accuracy gains and 30% lower inference cost when added to PPO or GRPO.
C3ot: Generating shorter chain-of-thought without compromising effectiveness
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
citing papers explorer
-
Self-Aligned Reward: Towards Effective and Efficient Reasoners
Self-aligned reward uses relative perplexity differences to encourage concise, query-specific reasoning in LLMs, yielding 4% accuracy gains and 30% lower inference cost when added to PPO or GRPO.
- Learning to Interrupt in Language-based Multi-agent Communication