PulseAugur
中
实时 18:21:13
English(EN) How to Teach Large Multimodal Models New Skills

新的微调方法教会大型多模态模型新技能且不遗忘

研究人员开发了新颖的方法来微调大型多模态模型(LMM),使其能够获得新技能而不会损害现有能力。通过分析输出令牌分布并识别“计数偏差”,他们发现仅选择性地更新自注意力投影层或MLP Gate&Up组件,可以显著减少先前学习任务上的性能损失。这些选择性微调方法在LLaVA-OneVision、LLaVA-NeXT和Qwen2.5-VL模型上进行了演示,在学习-稳定性权衡方面优于传统的全模型微调以及LwF、LoRA和WiSE-FT等其他遗忘缓解技术。 AI

影响 提供了更有效、更稳定的方法来使大型多模态模型适应新任务,而不会损害现有知识。

排序理由 该集群包含一篇详细介绍微调大型多模态模型新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

新的微调方法教会大型多模态模型新技能且不遗忘

本文如何被排名

Signal score
4 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇详细介绍微调大型多模态模型新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Zhen Zhu, Yiming Gong, Yao Xiao, Yaoyao Liu, Derek Hoiem ·

    How to Teach Large Multimodal Models New Skills

    arXiv:2510.08564v3 Announce Type: replace-cross Abstract: How can we teach large multimodal models (LMMs) new skills without erasing prior abilities? We study sequential fine-tuning on five target skills while monitoring general ability on eight held-out benchmarks across three m…