encoder

Tag

Cards List
#encoder

@maximelabonne: Train encoders today like it's 2020 again!

X AI KOLs Following · 4d ago Cached

Liquid AI fine-tuned their LFM2.5-Encoder models (230M and 350M) to perform multi-label classification in a single forward pass, eliminating the need for decoding loops or parsing. This demonstrates efficient label scoring for NLP tasks.

0 favorites 0 likes
#encoder

@vivmarquez: Not every AI problem is a generation problem. For routing, classification, retrieval, policy checks, and similar tasks,…

X AI KOLs Following · 5d ago Cached

Liquid AI releases LFM2.5-Encoder-230M and LFM2.5-Encoder-350M, bidirectional encoders optimized for non-generative tasks like classification and retrieval, offering fast CPU inference at long context.

0 favorites 0 likes
#encoder

@maximelabonne: We just released two new encoder models (MLM) in 2026 They're super fast, easy to train, and strongly multilingual. Try…

X AI KOLs Following · 6d ago Cached

Liquid AI released two new encoder models, LFM2.5-Encoder-230M and LFM2.5-Encoder-350M, that are fast, easy to train, and strongly multilingual, with speed benchmarks showing over 3.7x improvement on CPU compared to ModernBERT-base.

0 favorites 0 likes
#encoder

LFM2.5-Encoders for Fast Long-Context Inference on CPU

Hugging Face Blog · 6d ago Cached

Liquid AI releases LFM2.5-Encoders (230M and 350M), efficient encoder models optimized for long-context inference on CPU, matching or beating larger encoders on benchmarks with 3.7x speedup over ModernBERT-base.

0 favorites 0 likes
#encoder

LiquidAI/LFM2.5-Encoder-350M

Hugging Face Models Trending · 2026-07-27 Cached

Liquid AI releases LFM2.5-Encoder-350M, a multilingual bidirectional encoder built on the LFM2 architecture, offering strong quality for its size, 8k context, and efficient on-device performance across 15 languages.

0 favorites 0 likes
#encoder

Separating Representation from Reconstruction Enables Scalable Text Encoders

arXiv cs.CL · 2026-07-07 Cached

CrossBERT decouples representation learning from token reconstruction, enabling higher masking ratios and better sample efficiency, outperforming BERT on MTEB and GLUE benchmarks.

0 favorites 0 likes
#encoder

Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust Moderation

arXiv cs.AI · 2026-06-26 Cached

This paper introduces LeanGuard, a lightweight bidirectional encoder-based safety guardrail that matches the accuracy of larger reasoning-based guardrails while being approximately 100x faster, challenging the assumption that chain-of-thought reasoning is necessary for effective moderation.

0 favorites 0 likes
#encoder

Hellishly Slow Level 13 Deflate Compression

Hacker News Top · 2026-06-22 Cached

The article describes libdeflate's new level 13, a deliberately slow DEFLATE compression level that achieves marginally better compression (0.134% on Silesia) at the cost of being 56x slower than level 12, designed for scenarios where data is compressed once and decompressed many times.

0 favorites 0 likes
#encoder

m3BERT: A Modern, Multi-lingual, Matryoshka Bidirectional Encoder

arXiv cs.CL · 2026-05-20 Cached

This paper introduces m3BERT, a multilingual bidirectional encoder with a novel pretraining strategy that jointly optimizes representations across transformer layers and multiple embedding dimensions, enabling a single model to be adapted to varied resource constraints. It significantly outperforms state-of-the-art models on the Bing-Click industrial retrieval dataset.

0 favorites 0 likes
← Back to home

Submit Feedback