Knowledge catalogue
arXiv:2608.28478v1 Announce Type: new Abstract: Factual question answering (QA) typically assumes a single canonical answer, obscuring whether large language models (LLMs) retain divergent accounts of
arXiv:2608.27793v1 Announce Type: new Abstract: Autonomous navigation in underwater cave environments is essential for search-and-rescue operations, scientific exploration, and emergency egress. Tradi
arXiv:2608.28137v1 Announce Type: cross Abstract: We present CheXtriev, a graph-based, anatomy-aware framework for chest radiograph retrieval. Unlike prior methods focussed on global features, our met
Welcome to the second Cloud CISO Perspectives for August 2026. Today, Chris Sistrunk and Stephanie Kiel detail the critical issues facing the water sector, and actionable steps that OT operators can t
I've made a modification of llama.cpp MTP for people that want to run models like QWEN 27B on 16GB and similar setup, the focus is reducing the memory cost of MTP allowing more speed for less ctx cost
// ContextLeak in AI Agents // The whole attack surface here is a tool name and a tool description. Stealing an LLM agent's runtime context, meaning the user prompt, the execution trajectory and the t
arXiv:2608.28476v1 Announce Type: new Abstract: Long-horizon agentic tasks require large language models (LLMs) to iteratively retrieve, integrate, and maintain dispersed information across multi-turn
arXiv:2608.28547v1 Announce Type: new Abstract: Model merging combines multiple task-specific fine-tuned LLMs into a single multi-task model without additional training. However, merged models are kno
arXiv:2608.28302v1 Announce Type: new Abstract: Diffusion-based inpainting models modify only a localized part of an image, while many AI-image detectors rely on global artifacts and do not localize.
I keep seeing demos of AI agents building scenes in Blender through BlenderMCP, so I tried it myself. I ran both models locally for this and picked the GLM 5.3 family(Q4 quant) because videos of it do
I am the product lead for the training platform at @FireworksAI_HQ . Today the Training API and Fireworks Lab are GA. I won’t walk you through all the features. I want to tell you why teams train with
arXiv:2608.27866v1 Announce Type: new Abstract: Achieving virtual agents capable of automating tasks across diverse digital environments remains a pivotal challenge in Embodied AI. While Multimodal La
arXiv:2406.10221v3 Announce Type: replace-cross Abstract: Recent developments in vision-language models have significantly advanced video understanding. Existing datasets and tasks, however, have nota
Most agent tutorials stop before the hard parts. This one builds them step by step. Deep Agents from Scratch is a five-notebook LangGraph course for builders who want to understand how advanced AI age
My conversation with @jerryjliu0. He's building @llama_index: document infrastructure for AI agents, the layer underneath everything models need to work. Did you know he's the guy whose tool half the
When using kvarn quants at low context depth tg speed is similar to llama.cpp on and equivalent qx_x quant. However, as context depth grows kvarn tanks your tg speed. This fork optimises kvarn to have
arXiv:2512.05079v2 Announce Type: replace Abstract: Object geometry is key information for robot manipulation. Yet, object reconstruction is a challenging task because camera observations are partial
arXiv:2512.00234v3 Announce Type: replace Abstract: There has been significant progress in open-source text-only translation large language models (LLMs) with better language coverage and quality. How
arXiv:2608.28406v1 Announce Type: new Abstract: Human mistakes are inevitable when following instructions, yet they can lead to severe consequences. As such, there has been an increased interest in de
Mac Studio, 96GB unified memory (M3 Ultra). I want the largest usable Qwen3.8-Flash-Next setup, and I'd rather not burn 100GB of bandwidth on the wrong download. Here's my math. Please tell me which p
arXiv:2608.28444v1 Announce Type: new Abstract: Due to the nature of quadratic attention, Large Language Models (LLMs) consume a lot of memory and energy. Every new token costs more than the previous
arXiv:2509.19212v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) are increasingly deployed in real-world applications, yet their ability to make context-aware safety decisi
arXiv:2608.27948v1 Announce Type: new Abstract: Test-time adaptation (TTA) on graphs aims to adapt a graph neural network (GNN) that is well-trained on the training graph to the test graph, which invo
arXiv:2606.17399v2 Announce Type: replace Abstract: When small transformers grok modular multiplication, prior work reports that the learned embedding has a 'dense' Fourier spectrum requiring all freq
Today, we also open the door to Fireworks Lab: Embed forward-deployed researchers and engineers to co-design or build your model for you. Join leaders like @cursor_ai, @harvey, @figma, @vercel, @cogni
arXiv:2608.27514v1 Announce Type: new Abstract: Diffusion-based language models (dLLMs) enable parallel token generation through iterative denoising, but existing decoding strategies collapse to singl
arXiv:2608.28152v1 Announce Type: cross Abstract: Agitation fluctuates over short time horizons in people living with dementia, yet continuous physiological information for anticipating next-day risk
Together AI, in partnership with greptile, announces its AI‑powered code‑review platform that is trusted by more than 11,000 teams for reviewing pull requests within the full context of their codebase
The intelligence in my context is an aggregate index, I called the Agentic Coding Index, across most relevant agentic coding benchmarks: SWE-bench Pro, DeepSWE v1.1, Terminal-Bench (v4, v3, v2.1), Cod
Interesting paper on prompt optimization. They claim that a single-lineage prompt optimizer just matched GEPA on a smaller rollout budget. Prompt optimization has been drifting toward heavier machiner
I used the quantized n-gram to INT4, it's 32 GB, memory-mapped from disk. I confirmed that it works great on 150-160k context, and i was watching all the time my VRAM usage while doing single thread l
After a couple of days of faffing around, here's my current config in case it helps anyone out. Some of this is also valuable for a single Spark. AI slop below for you to read and paste into your own
If you own 4xR9700 and were waiting for the model to make them shine, then I have some good news for you! It's running at 80-120 tokens/second for generation and 12k token/second prefill for a single
I find both Qwen 3.8 27b and Qwen 3.8 Flash Next difficult to read. Here's some examples of what I mean: **Model-visible tool set per turn** (assembled by the host at provider-request time): persona t
I kept running into the same thing. Change a system prompt, get a different answer, and I couldn't tell if that was my change or just the sampler. The loop made it worse. Edit the Modelfile in another
happened simultaneously across the board. OpenCode, OpenRouter, and Ollama all ruined DeepSeek at the exact same moment. They literally turned a brilliant model into a complete idiot overnight. I abso
I wanted to share my successful setup for running a Qwen 3.8 27B model with a massive context window on a consumer 16GB GPU (RTX 4070 Ti SUPER). The goal was to fit everything into VRAM without sacrif
~20–30GB smaller than Unsloth/AesSedai Q4 at similar PPL. After several days of testing I released a set of mainline-compatible imatrix quants for Qwen3.8-Flash-Next. Goal: same quality band as the po
The official benchmark card for Ling-3.0-flash-Fin is a useful reminder that the unit being tested is rarely just “the model.” The release says most runs used temperature 1, top_p 0.95 and the highest
arXiv:2608.26236v1 Announce Type: new Abstract: We present a six-stage framework for auditing the reproducibility of scientific claims across a research literature within the computer science domain,
arXiv:2608.26194v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) systems have attracted significant interest for their ability to mitigate hallucinations in Large Language Models
arXiv:2608.27088v1 Announce Type: cross Abstract: While most phenotyping platforms rely primarily on image-based measurements, advanced plant characterization requires the integration of active physio
Atlassian uses Claude for complex Rovo workflows and Claude Code internally; its multimodel platform also uses Google Cloud.
arXiv:2607.18064v2 Announce Type: replace-cross Abstract: Coding agents can now be left alone to improve software against a score. In this pattern--recently popularized as 'autoresearch'--the agent re
arXiv:2608.27301v1 Announce Type: cross Abstract: In this study, we examined whether current 3D reconstruction methods can support the creation of realistic holographic representations of laboratory o
arXiv:2608.26465v1 Announce Type: new Abstract: Compositional generalization is usually evaluated through model accuracy. We instead ask which structural or lexical identifications make held-out COGS
arXiv:2608.27409v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) improves specific capabilities of large language models, but covering multiple capabilities often
arXiv:2608.27391v1 Announce Type: new Abstract: LLMs are increasingly able to answer complex questions about enterprise-scale document collections. But evaluation is hard: companies don't want to shar
arXiv:2608.26788v1 Announce Type: new Abstract: Recent work shows that pre-trained, instruction-tuned vision-language models (VLMs) perform well at mapping from instructions and observations to high-l
arXiv:2608.26674v1 Announce Type: cross Abstract: As large language models are increasingly deployed to simulate diverse human characters, ensuring persona fidelity, defined as the extent to which an
arXiv:2608.27110v1 Announce Type: new Abstract: Reliable medical conversational AI requires authentic expert--patient interaction data, yet such datasets remain scarce, especially for low-resource lan
arXiv:2608.26154v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly being integrated into digital health platforms to generate summaries of complex medical data. Although t
arXiv:2608.26110v1 Announce Type: cross Abstract: Large Language Models (LLMs) are emerging as promising assistants in High-Performance Computing (HPC), where programming remains complex and expertise
arXiv:2606.08357v2 Announce Type: replace Abstract: Diffusion language models generate text through iterative denoising, offering a powerful alternative to autoregressive generation. However, discrete
arXiv:2608.26117v1 Announce Type: cross Abstract: RDBMS (Relational Database Management System) databases face several limitations, including slow execution with multi-hop queries and a lack of explai
I built a benchmark called EchoNet. An agent gets a factual question, then searches a syntheic web I made before answering. Some of that web is seeded with misinformation: one fake page, a fake page r
arXiv:2608.26641v1 Announce Type: new Abstract: Decoding quality in diffusion multimodal language models (dMLLMs) depends heavily on the order in which masked tokens are committed. Existing confidence
arXiv:2608.26582v1 Announce Type: cross Abstract: Self-evolving language models have recently emerged as a promising path toward superintelligence, with the advantage of reducing the cost of human sup
Just a rumour of a bug is enough to find a security exploit these days Anil Madhavapeddy is a professor of computer science at Cambridge and a core maintainer of the OCaml compiler. In this somewhat a
arXiv:2608.26730v1 Announce Type: new Abstract: Large language models offer broad capabilities, but adapting them to evolving domains, tools, and requirements often entails repeated post-training. Aut