llm-kb
← Back to research
Paper

Researchers fine-tune SHuBERT-ByT5 for real-time sentence-level sign language translation

A team fine-tuned a SHuBERT-ByT5 translation stack on a subset of How2Sign using QLoRA, achieving a BLEU of 15.9 on the test set. The work focuses on real-time deployment rather than novel architectures, addressing the gap between isolated sign recognition and natural sentence-level translation.

0 engagement·1 source·Fri, Jul 10, 2026, 05:11 PM
The researchers used a uniformly sampled 9,872-example subset of How2Sign due to compute and storage constraints. They kept SHuBERT frozen and applied QLoRA to fine-tune the stack. Validation BLEU was 16.7, test BLEU 15.9, and BLEURT 44.7. The main contribution is demonstrating feasibility of real-time sentence-level SLT without proposing new architectures.

Entities

QLoRA(tool)SHuBERT(model)ByT5(model)How2Sign(benchmark)

Related

Paper

VTaMo introduces explicit multi-granularity alignment for sign language translation

VTaMo is a new framework for sign language translation that introduces explicit multi-granularity alignment between video and text at three levels: local alignment via optimal transport with a learnable null token, global alignment via orthogonal transformation, and sequence-level alignment. This approach aims to improve gloss-free SLT by providing more structured cross-modal correspondences.

0 engagement·1 source·arxiv
Fri, Jul 10, 2026, 06:30 AM
arXiv
Paper

Tokenizer Transplantation Fixes Autoregressive Collapse in Bengali ASR

Researchers propose a tokenizer transplantation pipeline to replace the English-centric byte-level tokenizer in Moonshine speech recognition models with the native-script BanglaBERT WordPiece vocabulary. This resolves catastrophic autoregressive collapse on Bengali, enabling edge-efficient ASR for morphologically rich non-Latin languages.

0 engagement·1 source·arxiv
Fri, Jul 10, 2026, 04:54 PM
arXiv
Paper

Researchers identify modality interference as root cause of full-duplex SLM degradation

A new paper on arXiv (July 7, 2026) presents a fine-grained analysis of optimization dynamics in full-duplex Spoken Language Models (SLMs), identifying severe modality interference as the root cause of knowledge degradation and compromised semantic integrity. The work aims to enable more natural and intelligent full-duplex SLMs.

0 engagement·1 source·arxiv
Tue, Jul 7, 2026, 05:43 PM
arXiv
Paper

Paper challenges text-only pretraining, proposes visual pretraining for language models

A new arXiv paper argues that current language model pretraining discards rich visual information from documents and web pages. The authors propose scalable visual pretraining to incorporate figures, equations, and layouts, aiming to improve language intelligence beyond text-only approaches.

0 engagement·1 source·arxiv
Fri, Jul 10, 2026, 05:57 PM
arXiv
Model Release

Soofi S 30B-A3B: Open-source MoE hybrid Mamba Transformer for German and English

Researchers released Soofi S 30B-A3B, a sovereign open-source Mixture-of-Experts foundation model for German and English. Its hybrid Mamba-Transformer design activates only 3B of 30B parameters per token, achieving throughput advantages for long-context deployment. Pretrained on 27 trillion tokens with up-weighted German data, it matches dense 14-27B models on English and German benchmarks while excelling in code tasks.

0 engagement·1 source·arxiv
Fri, Jul 10, 2026, 01:51 PM
arXiv