Fine-tuning Qwen2.5-3B with GRPO on GSM8K to use Prolog yields competitive zero-shot MMLU performance but exposes an accuracy-auditability trade-off interpreted as reward hacking.
Balancing exploration and exploitation in rl: A survey
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Training Language Models to Use Prolog as a Tool
Fine-tuning Qwen2.5-3B with GRPO on GSM8K to use Prolog yields competitive zero-shot MMLU performance but exposes an accuracy-auditability trade-off interpreted as reward hacking.