4 papers
Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding
Yuanhao Ding, Meimingwei Li, Esteban Garces Arias +3
In open-ended generation, LLMs frequently fall into the "likelihood trap", marked by repetitive degeneration and vocabulary dullness, creating a discrepancy between machine-generat…
Beyond Temperature: Hyperfitting as a Late-Stage Geometric Expansion
Meimingwei Li, Yuanhao Ding, Esteban Garces Arias +1
Recent work has identified a counterintuitive phenomenon termed "Hyperfitting", where fine-tuning Large Language Models (LLMs) to near-zero training loss on small datasets surprisi…
Min- Sampling: Decoupling Truncation from Temperature Scaling via Relative Logit Dynamics
Yuanhao Ding, Meimingwei Li, Esteban Garces Arias +3
The quality of text generated by large language models depends critically on the decoding sampling strategy. While mainstream methods such as Top-, Top-, and Min- achieve…
GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation
Yuanhao Ding, Esteban Garces Arias, Meimingwei Li +6
Open-ended text generation faces a critical challenge: balancing coherence with diversity in LLM outputs. While contrastive search-based decoding strategies have emerged to address…