PulseAugur
EN
LIVE 09:26:24
tool · [1 source] ·

New SFT objectives outperform NLL for capable LLMs

Researchers have explored alternative objectives for supervised fine-tuning (SFT) of large language models, moving beyond the standard negative log likelihood (NLL). Their study, involving extensive experiments across various models and benchmarks, reveals that different objectives perform better depending on the model's capability. Objectives that downweight low-probability tokens are more effective for highly capable models, while NLL excels with less capable models. AI

Summary written by gemini-2.5-flash-lite from 1 sources. How we write summaries →

IMPACT New fine-tuning objectives could improve LLM generalization and performance on specific tasks.

RANK_REASON The cluster contains an academic paper detailing new research findings on LLM fine-tuning objectives. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

COVERAGE [1]

  1. arXiv cs.CL TIER_1 · Gaotang Li, Ruizhong Qiu, Xiusi Chen, Heng Ji, Hanghang Tong ·

    Beyond Log Likelihood: Probability-Based Objectives for Supervised Fine-Tuning across the Model Capability Continuum

    arXiv:2510.00526v3 Announce Type: replace Abstract: Supervised fine-tuning (SFT) is the standard approach for post-training large language models (LLMs), yet it often shows limited generalization. We trace this limitation to its default training objective: negative log likelihood…