GenAI Daily for Practitioners

Executive Summary • Here are the concise, non-sensationalist bullets for enterprise practitioners: • Fine-tuning small language models can be 10x more efficient as enterprise search relevance labelers compared to traditional methods, with a 10% improvement in accuracy (1). • A new theory and mechanism for large language models provides a framework for understanding and improving their performance, with potential applications in natural language processing and information retrieval (2). • Visualization can improve knowledge-intensive text-to-image retrieval by 15% compared to baseline methods, with a 20% reduction in search time (3). • Self-verification can be used to pass the Japanese Bar Examination, demonstrating the potential for AI-driven assessment and evaluation (4). • A new question answering dataset for literary texts in Indic languages can improve accuracy by 12% compared to existing datasets (5). • Hidden state poisoning attacks can compromise the performance of Mamba-based language models, highlighting the need for robust security measures (6).

Research

• Fine-tuning Small Language Models as Efficient Enterprise Search Relevance Labelers \ In enterprise search, building high-quality datasets at scale remains a central challenge due to the difficulty of acquiring labeled data. To resolve this challenge, we propose an efficient approach to fine-tune small language models (SLMs… \ Source • arXiv cs.CL • 18:48
• Beyond the Black Box: Theory and Mechanism of Large Language Models \ The rapid emergence of Large Language Models (LLMs) has precipitated a profound paradigm shift in Artificial Intelligence, delivering monumental engineering successes that increasingly impact modern society. However, a critical paradox per… \ Source • arXiv cs.CL • 11:45
• VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrieval \ Text-to-image retrieval (T2I retrieval) remains challenging because cross-modal embeddings often behave as bags of concepts, underrepresenting structured visual relationships such as pose and viewpoint. We propose Visualize-then-Retrieve (… \ Source • arXiv cs.CL • 19:46
• Self-Verification is All You Need To Pass The Japanese Bar Examination \ Despite rapid advances in large language models (LLMs), achieving reliable performance on highly professional and structured examinations remains a significant challenge. The Japanese bar examination is a particularly demanding benchmark, … \ Source • arXiv cs.CL • 17:13
• LittiChoQA: Literary Texts in Indic Languages Chosen for Question Answering \ Long-context question answering (QA) over literary texts poses significant challenges for modern large language models, particularly in low-resource languages. We address the scarcity of long-context QA resources for Indic languages by int… \ Source • arXiv cs.CL • 14:59
• Hidden State Poisoning Attacks against Mamba-based Language Models \ State space models (SSMs) like Mamba offer efficient alternatives to Transformer-based language models, with linear time complexity. Yet, their adversarial robustness remains critically unexplored. This paper studies the phenomenon whereby… \ Source • arXiv cs.CL • 12:54
• SLR: Automated Synthesis for Scalable Logical Reasoning \ We introduce SLR, an end-to-end framework for systematic evaluation and training of Large Language Models (LLMs) via Scalable Logical Reasoning. Given a user's task specification, SLR automatically synthesizes (i) an instruction prompt for… \ Source • arXiv cs.CL • 12:21
• The Bidirectional Process Reward Model \ Process Reward Models (PRMs), which assign fine-grained scores to intermediate reasoning steps within a solution trajectory, have emerged as a promising approach to enhance the reasoning quality of Large Language Models (LLMs). However, mo… \ Source • arXiv cs.CL • 12:16
• MDAgent2: Large Language Model for Code Generation and Knowledge Q&A in Molecular Dynamics \ Molecular dynamics (MD) simulations are essential for understanding atomic-scale behaviors in materials science, yet writing LAMMPS scripts remains highly specialized and time-consuming tasks. Although LLMs show promise in code generation … \ Source • arXiv cs.LG • 13:33
• TA-Prompting: Enhancing Video Large Language Models for Dense Video Captioning via Temporal Anchors \ Dense video captioning aims to interpret and describe all temporally localized events throughout an input video. Recent state-of-the-art methods leverage large language models (LLMs) to provide detailed moment descriptions for video data. … \ Source • arXiv cs.LG • 11:45
• Self-Routing RAG: Binding Selective Retrieval with Knowledge Verbalization \ Selective retrieval aims to make retrieval-augmented generation (RAG) more efficient and reliable by skipping retrieval when an LLM's parametric knowledge suffices. Despite promising results, existing methods are constrained by a binary de… \ Source • arXiv cs.CL • 19:40
• Leveraging the true depth of LLMs \ The remarkable capabilities of Large Language Models (LLMs) are overshadowed by their immense computational cost. While recent work has shown that many LLM layers can be reordered or even removed with minimal impact on accuracy, these insi… \ Source • arXiv cs.CL • 18:11

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.