VCM is a training-free decoding intervention that applies PMI-driven token elevation and variance-adaptive penalization to reduce repetitive degeneration in LLM open-ended generation.
Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP , publisher =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2representative citing papers
citing papers explorer
-
Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding
VCM is a training-free decoding intervention that applies PMI-driven token elevation and variance-adaptive penalization to reduce repetitive degeneration in LLM open-ended generation.
- A Recipe for Long-Context Reasoning in Large Language Models via On-Policy Optimization and Distillation