RAG Without Vectors: How PageIndex Retrieves by Reasoning
avril 26, 2026admin NUAI,Committee,Actualités,Uncategorized(0)Retrieval is where most RAG systems quietly break. Traditional pipelines rely on vector similarity—embedding queries...
RAG vs. Context Stuffing: Why selective retrieval is more efficient and reliable than dumping all data into the prompt
février 24, 2026admin NUAI,Committee,Actualités,Uncategorized(0)Large context windows have dramatically increased how much information modern language models can process in...
RaDialog: A Large Vision-Language Model for Radiology Report Generation and Conversational Assistance
mai 8, 2025admin NUAI,Committee,Actualités,Uncategorized(0)arXiv:2311.18681v3 Announce Type: replace-cross Abstract: Conversational AI tools that can generate and discuss clinically correct...
RA3: Mid-Training with Temporal Action Abstractions for Faster Reinforcement Learning (RL) Post-Training in Code LLMs
octobre 9, 2025admin NUAI,Committee,Actualités,Uncategorized(0)TL;DR: A new research from Apple, formalizes what “mid-training” should do before reinforcement learning RL...
R-Zero: A Fully Autonomous AI Framework that Generates Its Own Training Data from Scratch
août 16, 2025admin NUAI,Committee,Actualités,Uncategorized(0)Large Language Models (LLMs) have revolutionized fields from natural language understanding to reasoning and code...
QwenLong-L1 solves long-context reasoning challenge that stumps current LLMs
mai 31, 2025admin NUAI,Committee,Actualités,Uncategorized(0)Alibaba’s QwenLong-L1 helps LLMs deeply understand long documents, unlocking advanced reasoning for practical enterprise applications.Read...
Qwen3-ASR-Toolkit: An Advanced Open Source Python Command-Line Toolkit for Using the Qwen-ASR API Beyond the 3 Minutes/10 MB Limit
septembre 20, 2025admin NUAI,Committee,Actualités,Uncategorized(0)Qwen has released Qwen3-ASR-Toolkit, an MIT-licensed Python CLI that programmatically bypasses the Qwen3-ASR-Flash API’s 3-minute/10...
Qwen Researchers Release Qwen3-TTS: an Open Multilingual TTS Suite with Real-Time Latency and Fine-Grained Voice Control
janvier 23, 2026admin NUAI,Committee,Actualités,Uncategorized(0)Alibaba Cloud’s Qwen team has open-sourced Qwen3-TTS, a family of multilingual text-to-speech models that target...
Quiet Feature Learning in Algorithmic Tasks
mai 8, 2025admin NUAI,Committee,Actualités,Uncategorized(0)arXiv:2505.03997v1 Announce Type: cross Abstract: We train Transformer-based language models on ten foundational algorithmic tasks...
