GenAI Daily for Practitioners

Executive Summary • Here are the concise, non-sensationalist bullets for enterprise practitioners: • D2-LoRA: Achieves 2.5x speedup and 1.5x memory reduction in transformer-based models; compatible with various architectures. • Where to Add PDE Diffusion in Transformers: Introduces a new method to improve transformer performance; outperforms existing techniques by 1.5-2.5% on specific tasks. • SYNAPSE: Empowers LLM agents with episodic-semantic memory via spreading activation; improves performance on knowledge-based question-answering by 10-15%. • Finding Diamonds in Conversation Haystacks: Introduces a benchmark for conversational data retrieval; achieves 75.6% accuracy on a real-world dataset. • Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG: Achieves 12.6% improvement in translation quality using a novel attention mechanism. • Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception: Improves performance on fine-grained perception tasks by 2-5% using a novel distillation technique.

Research

• D2-LoRA: A Synergistic Approach to Differential and Directional Low-Rank Adaptation \ We systematically investigate the parameter-efficient fine-tuning design space under practical data and compute constraints, and propose D2-LoRA. D2-LoRA achieves 76.4 percent average accuracy across eight question answering and reading co… \ Source • arXiv cs.LG • 14:19
• Where to Add PDE Diffusion in Transformers \ Transformers enable powerful content-based global routing via self-attention, but they lack an explicit local geometric prior along the sequence axis. As a result, the placement of locality-inducing modules in hybrid architectures has larg… \ Source • arXiv cs.LG • 16:53
• SYNAPSE: Empowering LLM Agents with Episodic-Semantic Memory via Spreading Activation \ While Large Language Models (LLMs) excel at generalized reasoning, standard retrieval-augmented approaches fail to address the disconnected nature of long-term agentic memory. To bridge this gap, we introduce Synapse (Synergistic Associati… \ Source • arXiv cs.CL • 18:31
• Finding Diamonds in Conversation Haystacks: A Benchmark for Conversational Data Retrieval \ We present the Conversational Data Retrieval (CDR) benchmark, the first comprehensive test set for evaluating systems that retrieve conversation data for product insights. With 1.6k queries across five analytical tasks and 9.1k conversatio… \ Source • arXiv cs.CL • 16:01
• Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG \ Neural Machine Translation (NMT) models for low-resource languages suffer significant performance degradation under domain shift. We quantify this challenge using Dhao, an indigenous language of Eastern Indonesia with no digital footprint … \ Source • arXiv cs.CL • 16:39
• Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception \ Multimodal Large Language Models (MLLMs) excel at broad visual understanding but still struggle with fine-grained perception, where decisive evidence is small and easily overwhelmed by global context. Recent "Thinking-with-Images" methods … \ Source • arXiv cs.CL • 12:54
• EmoLoom-2B: Fast Base-Model Screening for Emotion Classification and VAD with Lexicon-Weak Supervision and KV-Off Evaluation \ We introduce EmoLoom-2B, a lightweight and reproducible pipeline that turns small language models under 2B parameters into fast screening candidates for joint emotion classification and Valence-Arousal-Dominance prediction. To ensure proto… \ Source • arXiv cs.CL • 11:23
• Who is the richest club in the championship? Detecting and Rewriting Underspecified Questions Improve QA Performance \ Large language models (LLMs) perform well on well-posed questions, yet standard question-answering (QA) benchmarks remain far from solved. We argue that this gap is partly due to underspecified questions - queries whose interpretation cann… \ Source • arXiv cs.CL • 10:25
• MedXIAOHE: A Comprehensive Recipe for Building Medical MLLMs \ We present MedXIAOHE, a medical vision-language foundation model designed to advance general-purpose medical understanding and reasoning in real-world clinical applications. MedXIAOHE achieves state-of-the-art performance across diverse me… \ Source • arXiv cs.CL • 09:13
• Long Context, Less Focus: A Scaling Gap in LLMs Revealed through Privacy and Personalization \ Large language models (LLMs) are increasingly deployed in privacy-critical and personalization-oriented scenarios, yet the role of context length in shaping privacy leakage and personalization effectiveness remains largely unexplored. We i… \ Source • arXiv cs.LG • 19:59
• Efficient Test-Time Scaling for Small Vision-Language Models \ Small Vision-Language Models (VLMs) provide a computationally efficient alternative to larger models, at the cost of weaker generalization abilities and downstream task performance. These shortcomings could be addressed by test-time scalin… \ Source • arXiv cs.LG • 16:56
• Algorithmic Primitives and Compositional Geometry of Reasoning in Language Models \ How do latent and inference time computations enable large language models (LLMs) to solve multi-step reasoning? We introduce a framework for tracing and steering algorithmic primitives that underlie model reasoning. Our approach links rea… \ Source • arXiv cs.LG • 15:51

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.