Knowledge catalogue
arXiv:2608.21209v1 Announce Type: new Abstract: The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns. Prior work on contextual privacy studies whether LLM
I ran the Aider benchmark on Qwen 3.8 27B FP8 with FP8 KV cache 256K context vLLM. The score: 72.9 This matches Gemini 2.5 Pro from 2025-04-12 which also scored 72.9. Beats Claude Opus 4 from 2025-05-
With vision enabled I am able to run at 150k context on a single RTX 3090 and the results are just amazing. I was even able to write a gmail plugin for DeepSeek harness with locally hosted Qwen 3.8 27
arXiv:2512.00418v3 Announce Type: replace-cross Abstract: Significant Others (SOs) stabilize identity, regulate emotion, and support narrative meaning-making, yet many people today lack access to such
WARNING: I think I made a mistake with omlx, and its results below seem to be without any kind of acceleration. I am rerunning them with Lightning MTP and ANE support for prefill. Will update the post
Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies. A new model would arrive at the same price (or cheaper!) and paper over most of your problems.
Hello, So I've been trying lots of combinations in that never-ending landscape of options and settings. I wanted a proper quant of 3.8 27B running as fast as possible on my 5090 at 400W, with vision a
The latest RAG trend for the current agent harnesses (Codex, Cowork) is to do two passes of document processing to solve a knowledge work task over a data room of documents: 1️⃣ A fast and light pass,
Context: Earlier, open-source large models like Mimo V2.5 Pro, DeepSeek V4 Pro (first version), and Kimi K2.5 used to struggle with this prompt, and Qwen 3.6 27B couldn't even render the globe properl
Good afternoon, everyone! I wanted to show the work I've been doing around porting Ninfer over to the CMP170HX (Github) So, first, I do want to call out the amazing work that Neroued, Sergiuszm and sp
New/Old benchmark that provides a lot of answers for local LLM. I present to you a new test that I developed somewhat by accident: https://huggingface.co/cHunter789/Qwen3.8-27B-i1-IQ4_KS_KT-GGUF/tree/
Brace yourselves. We just entered a new era of frontier multimodal models. DeepSeek-V4-Flash-Vision-Exp advances multimodal agent performance. Also, pay attention to Ox Alpha 1M token contexts with te
arXiv:2608.19652v1 Announce Type: new Abstract: As LLM-based agents are deployed for longer and higher-stakes tasks, their memory systems continue to have crucial gaps. While existing memory benchmark
Welcome to the first Cloud CISO Perspectives for August 2026. Today, Chris Betz explains why the AI era makes it more important than ever to lean into security fundamentals.As with all Cloud CISO Pers
arXiv:2608.19950v1 Announce Type: cross Abstract: Artificial intelligence (AI) systems are increasingly used across domains to provide personalized information, recommendations, and decision support.
arXiv:2401.07468v4 Announce Type: replace-cross Abstract: The proposed model, CarSpeedNet, estimates scalar vehicle speed from a window of three-axis smartphone acceleration, without gyroscope, wheel-
I recently saw someone complaining that opencode causes compactions way too early (with 30k tokens still left). I fixed this issue and I thought I'd share my config to see if it helps anyone. I know p
Input tokens are often a meaningful part of the cost of running Retrieval Augmented Generation (RAG) at scale. This post describes a query-aware context compression pattern on Amazon Bedrock: after re
arXiv:2608.18305v1 Announce Type: new Abstract: X-ray fluorescence (XRF) microscopy maps elemental distributions, while optical microscopy can provide complementary morphological context. Localizing X
AirLLM dramatically reduces inference memory usage, letting 70B large language models run on a single 4GB GPU card — without quantization, distillation, or pruning. You can even run 405B Llama 3.1 on
Learn how to add context-aware security monitoring to FHIR APIs using Amazon Bedrock. This post shows how to detect anomalous access patterns, classify data sensitivity automatically, and generate com
arXiv:2608.18590v1 Announce Type: new Abstract: Compact object detectors are suitable for resource-constrained visual perception, but their limited representation capacity creates an accuracy gap rela
arXiv:2608.18581v1 Announce Type: cross Abstract: Although Large Language Models (LLMs) encode rich factual knowledge in their parameters, reliably recalling and verifying such knowledge remains a key
arXiv:2608.18968v1 Announce Type: new Abstract: Localized image edits can change a photograph's meaning while leaving most of it authentic, so forensic analysis must identify where an edit occurred. W
We are thrilled to announce that Google has been recognized as a Leader for the third year in a row in the 2026 Gartner® Magic Quadrant™ for Cloud-Native Application Platforms (CNAP). We believe this
Seeing a hype wave around GEN-1.5, and rightfully so. Lots of respect to Pete & Andy for executing so well. The secret is in the naturally repetitive motions in human-collected data. There're 2 main s
Very interesting new work Amazon and colleagues. (bookmark it) This connects to the emerging theme of treating the corpus as part of the harness. Planning loops, tool protocols, and context management
arXiv:2602.19816v3 Announce Type: replace-cross Abstract: For computational efficiency, modern language models are typically trained on independently sampled fixed-length sequences. Symbolic music lan
arXiv:2608.18661v1 Announce Type: new Abstract: Streaming text-to-speech is essential for low-latency spoken dialogue systems, yet many systems wait for sentence-level text and are therefore only pseu
arXiv:2608.17919v1 Announce Type: cross Abstract: Background and Context: Question and inquiry are integral parts of knowledge seeking and learning. Despite their importance, students tend not to ask
arXiv:2608.17911v1 Announce Type: new Abstract: As LLM agents operate across structured workflows and sessions, preserving long-term history does not ensure that later contexts can recover relevant ev
arXiv:2608.16919v1 Announce Type: cross Abstract: Recent advances in large language models and agent-based recommendation frameworks have introduced new opportunities for more flexible and context-awa
For over a decade, the Apache Hive Metastore (HMS) has served as the de facto metadata authority for big data analytics. Whether it was deployed on Hadoop clusters, self-managed Compute Engine VMs bac
Hello everyone, tried so hard to optimize my config and finally I simply get up to 70 t/s with q6 variant. And wanted to share with you guys so that other people with the same setup can enjoy. Please
arXiv:2608.17393v1 Announce Type: new Abstract: Reinforcement learning for coding agents increasingly relies on long-running agent harnesses to manage tool integration, repository contexts, and execut
arXiv:2608.17605v1 Announce Type: cross Abstract: Conversational AI is moving beyond isolated text prompts toward sustained, multimodal interaction. In real conversations, users clarify goals, revise
arXiv:2608.17304v1 Announce Type: cross Abstract: Formal verification is a crucial technique for ensuring the functional correctness of hardware designs. In the context of property checking, a key cha
arXiv:2608.16984v1 Announce Type: cross Abstract: Recent monocular depth estimators achieve strong zero-shot generalization, yet often struggle to preserve fine-grained structures and object boundarie
arXiv:2608.17443v1 Announce Type: new Abstract: Knowledge Graph Reasoning (KGR) aims to discover latent facts by leveraging the structural evidence available in KGs, posing a challenge to the structur
arXiv:2608.17719v1 Announce Type: cross Abstract: Context: Software systems that depend on commercial large language model APIs must migrate to successor versions when vendors deprecate older models.
arXiv:2608.17275v1 Announce Type: cross Abstract: AI agents increasingly act rather than merely read: across the Model Context Protocol (MCP) ecosystem, the share of deployed tools that modify externa
Zep offers a knowledge graph for persistent user memory and shared organisational context through MCP, according to Zep.
arXiv:2406.13049v3 Announce Type: replace-cross Abstract: Personalized phishing is difficult to defend against because messages can be tailored to a target's work, interests, and social context. Large
arXiv:2608.16044v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) answers a question by retrieving passages from a vector store and trusting them as context, so anyone who can add
arXiv:2508.08879v3 Announce Type: replace-cross Abstract: The growing deployment of large language models (LLMs) across diverse cultural contexts necessitates a deeper understanding of models' hidden
arXiv:2608.15124v1 Announce Type: new Abstract: In contextual optimization, the decision-maker seeks optimal decisions to minimize a cost function, that varies based on observed features. This context
I know the front page is completely dominated by the Qwen3.8 27B. I know it's a fantastic model and everyone is rightfully discussing it, but I wanted to shift gears and talk about another model I’ve
arXiv:2608.15533v1 Announce Type: cross Abstract: Linear attention models eliminate the quadratic prefix computation and context-growing KV cache of softmax attention by replacing pairwise token inter
arXiv:2608.14613v1 Announce Type: new Abstract: Modern LLM-agent frameworks increasingly interoperate through standards such as Anthropic's Model Context Protocol (MCP) for agent-to-tool access and Go
arXiv:2608.14808v1 Announce Type: new Abstract: When a user question is underspecified, a capable model should recognize that its context is insufficient, identify the missing information, ask for it,
arXiv:2608.14952v1 Announce Type: cross Abstract: A structured world-state (entities, relations, context, and predictive cues) is designed to preserve prediction-critical content when perception degra
arXiv:2608.16295v1 Announce Type: new Abstract: AI coding agents need more than relevant snippets: they need business semantics, validation evidence, relations, and assurance that their context is cur
arXiv:2608.16447v1 Announce Type: new Abstract: Long-horizon embodied tasks require LLM agents to iteratively decompose high-level goals, revise plans in response to environmental feedback, and ground
arXiv:2608.15008v1 Announce Type: new Abstract: Memory is becoming core infrastructure for long-horizon LLM agents, yet existing evaluations offer limited guidance on which memory substrate, namely th
Two days ago I released a hyper-optimized Qwen3.8-27B inference engine for an RTX 3090 (82 tps single request, 672 peak) - yesterday's update took that to 99 tps single-user / ~1,000 tps at 64 concurr
In this post, we describe how AIDA works at a high level and how it helps address these challenges — grounding users in the right contracts, under the right legal context, and within the right access
arXiv:2608.15354v1 Announce Type: new Abstract: LLMs are increasingly used in morally sensitive contexts, yet it is unclear whether they apply ethical principles consistently across situations. A mode
arXiv:2608.15797v1 Announce Type: new Abstract: KV-cache eviction caps the memory cost of long reasoning traces but is inherently lossy because the model decodes from a partial view of its history. Un
arXiv:2603.24465v2 Announce Type: replace Abstract: Recent advances in large language models (LLMs) and LLM-based agents have substantially improved the capabilities of automated theorem proving. Howe
arXiv:2608.14767v1 Announce Type: cross Abstract: Automated vehicles must explain their decisions in ways that passengers can understand, monitor, and trust. Existing language-annotated driving datase