About
Research & Development
Unmasking Privacy Risks: How LLM Tokenizers Leak Training Data
TLDR: A new study reveals that Large Language Model (LLM) tokenizers, often overlooked, are vulnerable to Membership Inference Attacks (MIAs). Researchers developed five attack...
Research & Development
Enhancing LLM Performance for Business Conversation Summarization with Domain-Adaptive Pre-Training
TLDR: This research introduces DACP (Domain-Adaptive Continual Pre-Training), a self-supervised method to adapt smaller Large Language Models (LLMs) for summarizing real-world, noisy phone conversations....
Research & Development
A New Benchmark Reveals LLMs Struggle with Constrained Planning
TLDR: LEXICON is a novel benchmark designed to evaluate large language models (LLMs) on planning tasks that incorporate temporal constraints described in natural language....
Research & Development
Efficient Speech Synthesis: Introducing ECTSpeech for High-Quality One-Step Generation
TLDR: ECTSpeech is a new framework for efficient, high-quality one-step speech synthesis. It applies an Easy Consistency Tuning (ECT) strategy to a pre-trained diffusion...
Research & Development
Accelerating Safe Autonomous Driving Through Human-in-the-Loop Reinforcement Learning
TLDR: Human-Guided Distributional Soft Actor-Critic (H-DSAC) is a new human-in-the-loop reinforcement learning method designed for safe and efficient real-world autonomous driving. It integrates human...
Research & Development
Stopping AI’s Visual Thinking Drift in Video Understanding
TLDR: A new research paper introduces "Visual Thinking Drift," a phenomenon where Chain-of-Thought (CoT) reasoning in video AI leads to hallucinations and errors due...