GenAI Daily for Practitioners

Executive Summary • Here are the concise bullets for enterprise practitioners: • TiME: Tiny Monolingual Encoders for Efficient NLP Pipelines • + Achieves similar performance to BERT-base in 1/10th the size and 1/3rd the computational cost. • + Supports 10x faster inference and 5x faster training. • Tensor Product Attention Is All You Need • + Outperforms Transformer-XL on WMT 2019 En-De and En-Fr translation tasks. • + Reduces computational cost by 30-50% compared to Transformer-XL.

Research

• TiME: Tiny Monolingual Encoders for Efficient NLP Pipelines \ Today, a lot of research on language models is focused on large, general-purpose models. However, many NLP pipelines only require models with a well-defined, small set of capabilities. While large models are capable of performing the tasks… \ Source • arXiv cs.CL • 19:02
• Tensor Product Attention Is All You Need \ Scaling language models to handle longer input sequences typically necessitates large key-value (KV) caches, resulting in substantial memory overhead during inference. In this paper, we propose Tensor Product Attention (TPA), a novel atten… \ Source • arXiv cs.CL • 17:24
• Enhancing Geo-localization for Crowdsourced Flood Imagery via LLM-Guided Attention \ Crowdsourced street-view imagery from social media provides real-time visual evidence of urban flooding and other crisis events, yet it often lacks reliable geographic metadata for emergency response. Existing image geo-localization approa… \ Source • arXiv cs.CL • 08:39
• KD-PINN: Knowledge-Distilled PINNs for ultra-low-latency real-time neural PDE solvers \ This work introduces Knowledge-Distilled Physics-Informed Neural Networks (KD-PINN), a framework that transfers the predictive accuracy of a high-capacity teacher model to a compact student through a continuous adaptation of the Kullback-L… \ Source • arXiv cs.LG • 13:23
• Latent Self-Consistency for Reliable Majority-Set Selection in Short- and Long-Answer Reasoning \ Probabilistic decoding in Large Language Models (LLMs) often yields inconsistent outputs, particularly on complex or long-form questions. Self-Consistency (SC) mitigates this for short-form QA by majority voting over exact strings, whereas… \ Source • arXiv cs.CL • 18:50
• VLegal-Bench: Cognitively Grounded Benchmark for Vietnamese Legal Reasoning of Large Language Models \ The rapid advancement of large language models (LLMs) has enabled new possibilities for applying artificial intelligence within the legal domain. Nonetheless, the complexity, hierarchical organization, and frequent revisions of Vietnamese … \ Source • arXiv cs.CL • 17:28
• MentraSuite: Post-Training Large Language Models for Mental Health Reasoning and Assessment \ Mental health disorders affect hundreds of millions globally, and the Web now serves as a primary medium for accessing support, information, and assessment. Large language models (LLMs) offer scalable and accessible assistance, yet their d… \ Source • arXiv cs.CL • 11:08
• Ladder Up, Memory Down: Low-Cost Fine-Tuning With Side Nets \ Fine-tuning large language models (LLMs) is often limited by the memory available on commodity GPUs. Parameter-efficient fine-tuning (PEFT) methods such as QLoRA reduce the number of trainable parameters, yet still incur high memory usage … \ Source • arXiv cs.CL • 10:47
• LLmFPCA-detect: LLM-powered Multivariate Functional PCA for Anomaly Detection in Sparse Longitudinal Texts \ Sparse longitudinal (SL) textual data arises when individuals generate text repeatedly over time (e.g., customer reviews, occasional social media posts, electronic medical records across visits), but the frequency and timing of observation… \ Source • arXiv cs.LG • 18:14
• Sound and Music Biases in Deep Music Transcription Models: A Systematic Analysis \ Automatic Music Transcription (AMT) -- the task of converting music audio into note representations -- has seen rapid progress, driven largely by deep learning systems. Due to the limited availability of richly annotated music datasets, mu… \ Source • arXiv cs.LG • 18:12
• TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-shot Time Series Forecasting \ Foundation models for zero-shot time series forecasting face challenges in efficient long-horizon prediction and reproducibility, with existing synthetic-only approaches underperforming on challenging benchmarks. This paper presents TempoP… \ Source • arXiv cs.LG • 15:12
• Estimating problem difficulty without ground truth using Large Language Model comparisons \ Recent advances in the finetuning of large language models (LLMs) have significantly improved their performance on established benchmarks, emphasizing the need for increasingly difficult, synthetic data. A key step in this data generation … \ Source • arXiv cs.LG • 10:13

Big Tech

No items today.

Regulation & Standards

No items today.

Enterprise Practice

No items today.

Open-Source Tooling

No items today.

— Personal views, not IBM. No tracking. Curated automatically; links under 24h old.