About
Research & Development
Unlocking Process Mining Data: A New Bilingual Dataset for Text-to-SQL Translation
TLDR: The text2SQL4PM dataset is a new bilingual (Portuguese-English) benchmark for text-to-SQL translation in the process mining domain. Developed by Bruno Y. Yamate et...
Research & Development
Unlocking Deeper Meaning: Waveforms for Semantic Memory and Retrieval
TLDR: The research paper introduces a novel wave-based semantic memory system, ResonanceDB, which represents semantic patterns as complex-valued waveforms with both amplitude and contextual...
Research & Development
Unveiling the Inner Workings of AI Refusal Mechanisms
TLDR: This research introduces a three-stage pipeline using Sparse Autoencoders (SAEs) and Factorization Machines (FMs) to identify and understand the causal features behind large...
Research & Development
Simulating Daily Journeys: How AI Models Create Realistic Travel Diaries
TLDR: A new study introduces a Large Language Model (LLM) framework to generate individual travel diaries for transportation models. It creates realistic individual profiles...
Research & Development
Unmasking the Core Triggers of Hallucination in Gemma Language Models
TLDR: A new research paper investigates the intrinsic causes of hallucination in Gemma Large Language Models (LLMs), identifying five key "symbolic properties"—modifiers, named entities,...
Research & Development
VStyle: A New Benchmark for Teaching AI to Speak with Style and Emotion
TLDR: VStyle is a new bilingual benchmark (Chinese & English) designed to evaluate how well Spoken Language Models (SLMs) can adapt their speaking style...