TokenTiming: A Dynamic Alignment Method for Universal Speculative Decoding Model Pairs
10月 20, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2510.15545v1 Announce Type: new Abstract: Accelerating the inference of large language models (LLMs) has been...
TokenSmith: Streamlining Data Editing, Search, and Inspection for Large-Scale Language Model Training and Interpretability
7月 28, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2507.19419v1 Announce Type: new Abstract: Understanding the relationship between training data and model behavior during...
Tokenization, Fusion and Decoupling: Bridging the Granularity Mismatch Between Large Language Models and Knowledge Graphs
2月 28, 2026admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2602.22698v1 Announce Type: new Abstract: Leveraging Large Language Models (LLMs) for Knowledge Graph Completion (KGC)...
Token Buncher: Shielding LLMs from Harmful Reinforcement Learning Fine-Tuning
8月 29, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2508.20697v1 Announce Type: cross Abstract: As large language models (LLMs) continue to grow in capability...
Together AI’s ATLAS adaptive speculator delivers 400% inference speedup by learning from workloads in real-time
10月 11, 2025admin NUAI,Committee,ニュース,Uncategorized(0)Enterprises expanding AI deployments are hitting an invisible performance wall. The culprit? Static speculators that...
To Retrieve or To Think? An Agentic Approach for Context Evolution
1月 14, 2026admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2601.08747v1 Announce Type: new Abstract: Current context augmentation methods, such as retrieval-augmented generation, are essential...
Time Magazine appears to accidentally publish embargoed story confirming new Anthropic model
5月 22, 2025admin NUAI,Committee,ニュース,Uncategorized(0)Someone also appears to have published a full scrape of the Time article online on...
Tilde Research Introduces Aurora: A Leverage-Aware Optimizer That Fixes a Hidden Neuron Death Problem in Muon
5月 12, 2026admin NUAI,Committee,ニュース,Uncategorized(0)Researchers at Tilde Research have released Aurora, a new optimizer for training neural networks that...
Tilde AI Releases TildeOpen LLM: An Open-Source Large Language Model with Over 30 Billion Parameters and Support Most European Languages
9月 7, 2025admin NUAI,Committee,ニュース,Uncategorized(0)Latvian language-tech firm Tilde has released TildeOpen LLM, an open-source foundational large language model (LLM)...
