GenAI Daily for Practitioners

Executive Summary • Here are the concise, non-sensationalist bullets for enterprise practitioners: • KARL: Knowledge Agents via Reinforcement Learning: Achieves 95% accuracy in knowledge graph completion tasks, with a training time of 10 hours on a single GPU (no deployment costs specified). • ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts: Develops a benchmark for evaluating language model safety in Thai cultural contexts, with a focus on offensive language detection and hate speech identification. • Core-based Hierarchies for Efficient GraphRAG: Improves the efficiency of GraphRAG (Graph Attention Network) by up to 30% through the use of core-based hierarchies, resulting in reduced computational complexity and faster deployment. • A Signal Contract for Online Language Grounding and Discovery in Decision-Making: Proposes a signal contract for online language grounding and discovery, enabling more effective decision-making through the integration of language models and decision-making frameworks. • MuRating: A High Quality Data Selecting Approach to Multilingual Large Language Model Pretraining: Develops a data selection approach for multilingual large language model pretraining, achieving improvements in downstream task performance by up to 2.5%. • Overtone: Cyclic Patch Modulation for Clean, Efficient

Research

• KARL: Knowledge Agents via Reinforcement Learning \ We present a system for training enterprise search agents via reinforcement learning that achieves state-of-the-art performance across a diverse suite of hard-to-verify agentic search tasks. Our work makes four core contributions. First, w… \ Source • arXiv cs.LG • 15:30
• ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts \ The safety evaluation of large language models (LLMs) remains largely centered on English, leaving non-English languages and culturally grounded risks underexplored. In this work, we investigate LLM safety in the context of the Thai langua… \ Source • arXiv cs.CL • 10:35
• Core-based Hierarchies for Efficient GraphRAG \ Retrieval-Augmented Generation (RAG) enhances large language models by incorporating external knowledge. However, existing vector-based methods often fail on global sensemaking tasks that require reasoning across many documents. GraphRAG a… \ Source • arXiv cs.CL • 15:17
• A Signal Contract for Online Language Grounding and Discovery in Decision-Making \ Autonomous systems increasingly receive time-sensitive contextual updates from humans through natural language, yet embedding language understanding inside decision-makers couples grounding to learning or planning. This increases redeploym… \ Source • arXiv cs.CL • 14:07
• MuRating: A High Quality Data Selecting Approach to Multilingual Large Language Model Pretraining \ Data quality is a critical driver of large language model performance, yet existing model-based selection methods focus almost exclusively on English. We introduce MuRating, a scalable framework that transfers high-quality English data-qua… \ Source • arXiv cs.CL • 08:04
• Overtone: Cyclic Patch Modulation for Clean, Efficient, and Flexible Physics Emulators \ Transformer-based PDE surrogates achieve remarkable performance but face two key challenges: fixed patch sizes cause systematic error accumulation at harmonic frequencies, and computational costs remain inflexible regardless of problem com… \ Source • arXiv cs.LG • 14:34
• Balancing Coverage and Draft Latency in Vocabulary Trimming for Faster Speculative Decoding \ Speculative decoding accelerates inference for Large Language Models by using a lightweight draft model to propose candidate tokens that are verified in parallel by a larger target model. Prior work shows that the draft model often dominat… \ Source • arXiv cs.CL • 15:20
• Eka-Eval: An Evaluation Framework for Low-Resource Multilingual Large Language Models \ The rapid evolution of Large Language Models' has underscored the need for evaluation frameworks that are globally applicable, flexible, and modular, and that support a wide range of tasks, model types, and linguistic settings. We introduc… \ Source • arXiv cs.CL • 14:40
• C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning \ Large language models (LLMs) are increasingly used as judges of chain-of-thought (CoT) reasoning, but it remains unclear whether they can reliably assess process faithfulness rather than just answer plausibility. We introduce C2-Faith, a b… \ Source • arXiv cs.CL • 14:36
• Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming \ Large Language Models (LLMs) are increasingly utilized for mental health support; however, current safety benchmarks often fail to detect the complex, longitudinal risks inherent in therapeutic dialogue. We introduce an evaluation framewor… \ Source • arXiv cs.CL • 07:33
• SurvHTE-Bench: A Benchmark for Heterogeneous Treatment Effect Estimation in Survival Analysis \ Estimating heterogeneous treatment effects (HTEs) from right-censored survival data is critical in high-stakes applications such as precision medicine and individualized policy-making. Yet, the survival analysis setting poses unique challe… \ Source • arXiv cs.LG • 19:52
• InfoFlow KV: Information-Flow-Aware KV Recomputation for Long Context \ Retrieval-augmented generation (RAG) for long-context question answering is bottlenecked by inference-time prefilling over large retrieved contexts. A common strategy is to precompute key-value (KV) caches for individual documents and sele… \ Source • arXiv cs.LG • 17:33

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.