Knowledge catalogue
arXiv:2604.06596v1 Announce Type: cross Abstract: Semi-supervised learning aims to infer class labels using only a small fraction of labeled data. In graph-based semi-supervised learning, this is typi
arXiv:2604.07003v1 Announce Type: new Abstract: Large language models (LLMs) has been widely used for automated negotiation, but their high computational cost and privacy risks limit deployment in pri
arXiv:2604.07350v1 Announce Type: cross Abstract: Large Chunk Test-Time Training (LaCT) has shown strong performance on long-context 3D reconstruction, but its fully plastic inference-time updates rem
arXiv:2604.07084v1 Announce Type: cross Abstract: Open-loop end-to-end neural motion planners have recently been proposed to improve motion planning for robotic manipulators. These methods enable plan
arXiv:2604.08131v1 Announce Type: new Abstract: The rapid spread of online misinformation has led to increasingly complex detection models, including large language models and hybrid architectures. Ho
arXiv:2604.08048v1 Announce Type: new Abstract: Classifier-Free Guidance (CFG) is a widely used inference-time technique to boost the image quality of diffusion models. Yet, its reliance on text condi
arXiv:2604.07937v1 Announce Type: new Abstract: Cross-document relation extraction (RE) aims to identify relations between the head and tail entities located in different documents. Existing approache
The ASUS ROG Flow Z13 (2025) with AMD Ryzen AI Max+ 395 (Strix Halo) and 128GB of unified LPDDR5X memory is a capable portable option for running large LLMs locally, with ASUS officially stating it...
arXiv:2604.06691v1 Announce Type: new Abstract: Real world deployment of multi agent reinforcement learning MARL systems is fundamentally constrained by limited compute memory and inference time. Whil
LiteParse hit 4K+ GitHub stars in 3 weeks. ~500 pages in 2 seconds. No GPU. No API keys. 50+ file formats. Now @LoganMarkewich, our Head of Open Source, will show you how to build with it. Live worksh
arXiv:2604.07823v1 Announce Type: new Abstract: Performance, the externalization of intent, emotion, and personality through visual, vocal, and temporal behavior, is what makes a character alive. Lear
The Mac Mini M4 Pro with 48GB unified memory is a popular choice in the local AI community for running large language models via Ollama, as its Apple Silicon architecture makes all 48GB of RAM dire...
arXiv:2604.07345v1 Announce Type: cross Abstract: The rapid growth of generative artificial intelligence (AI) has introduced unprecedented computational demands, driving significant increases in the e
arXiv:2604.06505v1 Announce Type: cross Abstract: Large language models (LLMs) are widely explored for reasoning-intensive research tasks, yet resources for testing whether they can infer scientific c
arXiv:2604.06881v1 Announce Type: new Abstract: Neural operators have emerged as powerful surrogates for dynamical systems due to their grid-invariant properties and computational efficiency. However,
arXiv:2604.03336v2 Announce Type: replace Abstract: BitNet b1.58 (Ma et al., 2024) demonstrates that large language models can operate entirely on ternary weights {-1, 0, +1}, yet no native binary wir
arXiv:2604.08503v1 Announce Type: new Abstract: Recent advances in generative video modeling, driven by large-scale datasets and powerful architectures, have yielded remarkable visual realism. However
arXiv:2604.04988v1 Announce Type: cross Abstract: Modern deployment often requires trading accuracy for efficiency under tight CPU and memory constraints, yet common compression proxies such as parame
arXiv:2604.05704v2 Announce Type: replace Abstract: Multimodal Sentiment Analysis (MSA) aims to infer human sentiment from textual, acoustic, and visual signals. In real-world scenarios, however, mult
arXiv:2604.07939v1 Announce Type: new Abstract: In this work, we present RAGE-XY, an extended version of RAGE, a real-time estimation framework that simultaneously infers vehicle velocity, tire slip a
arXiv:2604.07102v1 Announce Type: cross Abstract: Activation-based steering can personalize large language models at inference time, but its effects in educational settings remain unclear. We study pe
The Information wrote that Meta consumes 60.2T tokens per month, or ~750M per employee. Meta is getting absolutely token-mogged by us currently. SemiAnalysis employees consume 1.86B tokens per month.
arXiv:2604.06896v1 Announce Type: new Abstract: Epithelial tissues dynamically reshape through local mechanical interactions among cells, a process well captured by vertex models. Yet their many tunab
llama.cpp release **b8739** is a build of the open-source C/C++ LLM inference engine that introduces HIP backend support for the CDNA4 (gfx950) GPU architecture, enabling hardware acceleration on A...
Training LLMs requires periodic checkpoints — full snapshots of model weights, optimizer states, and gradients — whose storage costs can reach $200,000/month for a 405B model on 128 NVIDIA DGX B200...
Local AI tools such as Ollama, LM Studio, and Jan all rely on the same underlying inference engine (llama.cpp) and support compatible model formats (primarily GGUF), meaning a single downloaded mod...
If you’ve encountered garbled output like this while using GLM-5 or GLM-5.1 on our official service, the issue is now resolved. We've patched the underlying inference-side bugs and will be releasing a
With Amazon Bedrock Projects, you can attribute inference costs to specific workloads and analyze them in AWS Cost Explorer and AWS Data Exports. In this post, you will learn how to set up Projects en
Basalt: Qwen 3.8 FN NVFP4, single 5090 Hi guys, so I finally got a RAM upgrade and have been working on the top two pain points from Basalt: NVFP4 support and faster prefill. NVFP4: Basalt runs NVFP4
Got lucky with a 5070 Ti for $750 AFTER TAX (might be the best decision I've ever made tbh) back in November 2025. If only I had known I'd be getting into local llms at the time...anyways. I recently
With winter coming up where I'm at, I wanted to see if this could actually work. Running my 4x V620 server flat out made my apartment too hot anyway, so the problem really was control. I settled on us
Lots of hype around Strata lately and the speeds look great. But everything I see is tokens per second and needle tests. Has anyone checked if quality holds up? Same GGUF on llama.cpp should answer th
arXiv:2507.22612v3 Announce Type: replace-cross Abstract: Speech-to-text alignment is a critical component of neural text to speech (TTS) models. Autoregressive TTS models typically use an attention m
arXiv:2610.20841v1 Announce Type: cross Abstract: Many resource-constrained edge devices require artificial intelligence (AI) services but are not powerful enough to store or execute neural networks (
arXiv:2610.21763v1 Announce Type: new Abstract: Audio-enabled large language models (AuLLMs) allow users to interact with them simply by talking to them, without the need for an intermediate transcrip
arXiv:2610.22167v1 Announce Type: new Abstract: Understanding a conversation requires connecting what is said to the speakers' relationships, expectations, and earlier exchanges. Prior work shows that
arXiv:2610.21569v1 Announce Type: new Abstract: Diffusion models achieve strong performance in text-to-image generation, but they can still produce unsafe content. Training-free safety guidance offers
arXiv:2610.22340v1 Announce Type: cross Abstract: How can robots autonomously collect the contact-rich experience needed to learn dexterous manipulation? We present DSP: Dexterity from supervised play
arXiv:2610.21347v1 Announce Type: new Abstract: Large Language Models (LLMs) increasingly operate in multi-turn interactions that require maintaining latent conversational states, not only explicit fa
arXiv:2610.22127v1 Announce Type: cross Abstract: Personalization of Large Language Models (LLMs) is increasingly important for user-centric applications, yet prior methods mainly adapt dense LLMs, wh
arXiv:2610.22061v1 Announce Type: new Abstract: Organizations increasingly encode operational procedures as multi-step agent skills, here called Agent Operating Procedures (AOPs), that must be traceab
arXiv:2610.21167v1 Announce Type: new Abstract: Deep research agents are increasingly deployed to retrieve precise information from large corpora, yet multi-hop, long-horizon tasks remain challenging.
arXiv:2610.20681v1 Announce Type: new Abstract: Formal inequality proving requires mathematical reasoning and exact computation, often connected through intermediate lemmas and goal reformulations. We
arXiv:2610.20818v1 Announce Type: new Abstract: Unified multimodal generation covers discrete, continuous, and hybrid paradigms. Discrete approaches with single-codebook visual tokenizers align natura
arXiv:2610.21411v1 Announce Type: new Abstract: Multi-agent LLM systems rely on generative models for expert execution, workflow planning, and artifact review. Yet many coordination operations require
arXiv:2603.22519v4 Announce Type: replace-cross Abstract: Textual Large Language Models (LLMs) provide a simple and familiar interface: a string of text is used for both input and output. However, the
arXiv:2610.21630v1 Announce Type: new Abstract: Time series such as recordings of human motion data often encode structure across multiple temporal scales, and when a latent state determining this seq
arXiv:2610.21362v1 Announce Type: cross Abstract: Robotic manipulation often requires remembering interactions that are no longer visible. We introduce Latent Transition Memory (LTM), which stores his
arXiv:2610.21300v1 Announce Type: new Abstract: To show that a medical device complies with regulation, experts judge whether documents and records meet each requirement and record the judgment for ot
arXiv:2610.21375v1 Announce Type: cross Abstract: Memory enables large language models (LLMs) to use information beyond their parameters, such as past user conversations, long documents, or an agent's
arXiv:2610.21185v1 Announce Type: new Abstract: Modern generative models produce high-quality individual outputs but struggle to sample diversely, even when explicitly instructed to do so. We introduc
arXiv:2610.22193v1 Announce Type: new Abstract: In robotic manipulation, the same observation often admits multiple feasible actions. Generative policies such as diffusion and flow matching naturally
arXiv:2606.19911v2 Announce Type: replace Abstract: Decentralized deployment of LLM agents with diverse capabilities motivates infrastructure for knowledge sharing across heterogeneous agent populatio
arXiv:2610.21125v1 Announce Type: new Abstract: Few-step generative models produce high-quality samples in one or a few evaluations. However, steering them toward desired properties remains difficult,
arXiv:2610.22092v1 Announce Type: new Abstract: Aerial robots operating near people must treat a pedestrian, a parked car, and a wall differently, yet conventional perception--planning stacks reduce a
arXiv:2507.09227v2 Announce Type: replace-cross Abstract: Interest in generating realistic synthetic medical images has grown in recent years. Such datasets can mitigate the scarcity of public data fo
arXiv:2610.21308v1 Announce Type: cross Abstract: Consumer phones and wearables increasingly collect physiological time-series data, yet most mobile systems treat these streams as summaries or text ra
arXiv:2610.21390v1 Announce Type: cross Abstract: Multi-agent large language model (LLM) systems are increasingly deployed at the edge to support complex tasks that require collaboration and informati
arXiv:2610.22297v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) enhances the capabilities of large language models (LLMs) by querying external structured knowledge. However, it ca
arXiv:2610.21333v1 Announce Type: new Abstract: Natural language messages often contain quasi-identifiers that are weak in isolation but can jointly reveal an individual's private attributes. We call