Escaping the Mode Lottery: Multi-Response Training Improves Language Model Generalization 事件

Escaping the Mode Lottery: Multi-Response Training Improves Language Model Generalization arXiv:2606.00544v1 Announce Type: cross Abstract: Modern language-model fine-tuning typically pairs each prompt with a single response, even though many prompts admit multiple valid completions. This effectively reduces a multi-modal conditional distribution to a one-sample view, a phenomenon we call the "mode lottery," where training emphasizes a subset of plausible modes while leaving others underreprese