PulseAugur
EN
LIVE 14:56:18

New framework enhances LLM training by reducing noise in weaker models

By PulseAugur Editorial · [3 sources] ·

Researchers have developed a new framework called Contrastive Weak-to-Strong Generalization (ConG) to improve the training of large language models. ConG addresses limitations in existing weak-to-strong generalization methods, which can be hindered by noise and biases from weaker models. By leveraging implicit rewards and contrastive decoding, ConG generates higher-quality samples, leading to more reliable capability transfer and improved robustness. This approach has demonstrated consistent improvements across various model families, offering a promising path for advancing LLM training. AI

IMPACT Enhances LLM training by improving sample quality and robustness, potentially accelerating progress towards more capable models.

RANK_REASON The cluster contains two academic papers detailing novel methods for improving large language model training.

Read on Hugging Face Daily Papers →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

New framework enhances LLM training by reducing noise in weaker models

COVERAGE [3]

  1. arXiv cs.AI TIER_1 English(EN) · Houcheng Jiang, Junfeng Fang, Jiaxin Wu, Tianyu Zhang, Chen Gao, Xiang Wang, Xiangnan He, Yang Deng ·

    Contrastive Weak-to-strong Generalization

    arXiv:2510.07884v2 Announce Type: replace-cross Abstract: Weak-to-strong generalization provides a promising paradigm for scaling large language models (LLMs) by training stronger models on samples from aligned weaker ones, without requiring human feedback or explicit reward mode…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    Weak-to-Strong Generalization via Direct On-Policy Distillation

    Direct On-Policy Distillation transfers reinforcement learning improvements from smaller to larger models by using the policy shift induced by RL as an implicit reward signal, enabling efficient scaling of training without re-running expensive RL on the target model.

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    Weak-to-Strong Generalization via Direct On-Policy Distillation

    Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training. As models scale, post-training itself b…

RELATED ENTITIES

RELATED TOPICS