AI News AI资讯 4h ago Updated 2h ago 更新于 2小时前 46

Liquid AI Releases LFM2.5-Encoder-230M and LFM2.5-Encoder-350M: Bidirectional Encoders That Stay Fast at 8K Context on CPU Liquid AI 发布 LFM2.5-Encoder-230M 和 LFM2.5-Encoder-350M:在 8K 上下文中保持快速的 bidirectional 编码器

Liquid AI released two open-weight bidirectional encoders, LFM2.5-Encoder-230M and LFM2.5-Encoder-350M, built on the LFM2 hybrid backbone with an 8,192-token context window. The models are converted from decoder backbones by replacing causal attention with bidirectional masking, making convolutions non-causal, and training with a 30% masked language modeling objective. LFM2.5-Encoder-350M ranks fourth among 14 models on a 17-task benchmark (mean score 81.02), outperforming larger models like XLM Liquid AI发布了两个开源双向编码器LFM2.5-Encoder-230M和LFM2.5-Encoder-350M,它们基于LFM2混合骨干架构,拥有8,192个token的上下文窗口。这些模型通过将解码器骨干中的因果注意力替换为双向掩码、使卷积变为非因果,并使用30%掩码语言建模目标进行训练而得到。在包含14个模型的17项任务基准测试中,LFM2.5-Encoder-350M排名第四(平均得分81.02),其效率与性能之比优于XLM-R XL和ModernBERT-large等更大规模的模型。这两个编码器均针对仅CPU部署进行了优化,推理速度显著快于竞争对手——例如,在CPU上处理8K

65
Hot 热度
70
Quality 质量
60
Impact 影响力

Analysis 深度分析

TL;DR

  • Liquid AI released two open-weight bidirectional encoders, LFM2.5-Encoder-230M and LFM2.5-Encoder-350M, built on the LFM2 hybrid backbone with an 8,192-token context window.
  • The models are converted from decoder backbones by replacing causal attention with bidirectional masking, making convolutions non-causal, and training with a 30% masked language modeling objective.
  • LFM2.5-Encoder-350M ranks fourth among 14 models on a 17-task benchmark (mean score 81.02), outperforming larger models like XLM-R XL and ModernBERT-large in efficiency-to-performance ratio.
  • Both encoders are optimized for CPU-only deployment, achieving significantly faster inference than competitors—e.g., ~28s vs. >90s for ModernBERT-base at 8K tokens on CPU.
  • Designed for edge devices, regulated industries (finance, healthcare, legal), and high-volume pipelines requiring low-latency, long-context encoding without GPU reliance.

Why It Matters

These encoders address a critical gap in production AI systems: efficient, accurate, long-context processing on CPUs where GPUs are unavailable or cost-prohibitive. Their performance relative to size makes them ideal for real-time applications such as document analysis, compliance filtering, and intent routing in constrained environments. As organizations increasingly deploy NLP models at scale under strict latency and hardware budgets, Liquid AI’s approach offers a practical path forward without sacrificing accuracy or context length.

Technical Details

  • Architecture: Based on LFM2 hybrid backbone combining gated short-convolution blocks with grouped-query attention; hidden size = 1024, vocabulary size = 65,536, supports 15 languages.
  • Conversion Method: Transformed from LFM2.5 decoders via three key changes: (1) causal → bidirectional attention mask, (2) short convolutions made non-causal using symmetric center padding, (3) trained with 30% MLM rate (vs. BERT’s 15%).
  • Training Pipeline: Two-stage process—first stage uses 1K-token context on large web corpus for general competence; second stage extends to 8K tokens across full data mix to enhance factual, legal, and multilingual understanding.
  • Benchmarking: Evaluated on 17 tasks from GLUE, SuperGLUE, and multilingual classification; all models fine-tuned per task using identical AdamW recipe from EuroBERT card, with learning rates selected across 10 values over 3 seeds, final scores averaged over 5 fresh seeds.
  • Deployment Optimization: Designed for CPU inference; demonstrated ~28s forward pass at 8K tokens for 230M model vs. >90s for ModernBERT-base; supports Hugging Face transformers integration with trust_remote_code=True.

Industry Insight

The release signals a growing industry shift toward lightweight, long-context encoders tailored for edge and regulated deployments where cloud dependency is unacceptable or too costly. Companies building safety filters, PII detectors, or document classifiers can now adopt models that balance speed, context length, and accuracy without requiring expensive GPU infrastructure. Additionally, the open-sourced evaluation methodology sets a new standard for fair, reproducible benchmarking of encoder architectures—encouraging transparency and accelerating innovation in efficient NLP components.

摘要

Liquid AI发布了两个开源双向编码器LFM2.5-Encoder-230M和LFM2.5-Encoder-350M,它们基于LFM2混合骨干架构,拥有8,192个token的上下文窗口。这些模型通过将解码器骨干中的因果注意力替换为双向掩码、使卷积变为非因果,并使用30%掩码语言建模目标进行训练而得到。在包含14个模型的17项任务基准测试中,LFM2.5-Encoder-350M排名第四(平均得分81.02),其效率与性能之比优于XLM-R XL和ModernBERT-large等更大规模的模型。这两个编码器均针对仅CPU部署进行了优化,推理速度显著快于竞争对手——例如,在CPU上处理8K tokens时,ModernBERT-base需要超过90秒,而只需约28秒。该设计面向边缘设备、受监管行业(金融、医疗、法律)以及需要低延迟长上下文编码且无需GPU支持的高吞吐量管道。

深入分析

TL;DR

  • Liquid AI发布了两个开源双向编码器LFM2.5-Encoder-230M和LFM2.5-Encoder-350M,它们基于LFM2混合骨干架构,拥有8,192个token的上下文窗口。
  • 这些模型通过将解码器骨干中的因果注意力替换为双向掩码、使卷积变为非因果,并使用30%掩码语言建模目标进行训练而得到。
  • 在包含14个模型的17项任务基准测试中,LFM2.5-Encoder-350M排名第四(平均得分81.02),其效率与性能之比优于XLM-R XL和ModernBERT-large等更大规模的模型。
  • 这两个编码器均针对仅CPU部署进行了优化,推理速度显著快于竞争对手——例如,在CPU上处理8K tokens时,ModernBERT-base需要超过90秒,而只需约28秒。
  • 该设计面向边缘设备、受监管行业(金融、医疗、法律)以及需要低延迟长上下文编码且无需GPU支持的高吞吐量管道。

为什么重要

这些编码器解决了生产AI系统中的一个关键问题:在没有GPU或GPU成本过高的情况下,在CPU上实现高效、准确的长上下文处理。相对于其规模而言的性能表现使它们非常适合实时应用,如文档分析、合规性过滤和受限环境中的意图路由。随着组织越来越多地在严格的延迟和硬件预算下大规模部署NLP模型,Liquid AI的方法提供了一种切实可行的前进路径,同时不牺牲准确性或上下文长度。

技术细节

  • 架构:基于LFM2混合骨干架构组合

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Open Source 开源 Embedding Model 嵌入模型 Inference 推理 Research 科学研究