Knowledge catalogue
LightOn OCR-3 is now available on OpenDocRouter, priced at 0.28 / 1m input and 1.40 / 1m output (it is ~$3.19 / 1k pages on ParseBench). We benchmarked it on ParseBench. It is on the pareto frontier f
arXiv:2604.01569v2 Announce Type: replace Abstract: Video multimodal large language models achieve strong results on existing benchmarks, but answer accuracy alone does not establish whether they can
Yesterday TechCrunch published a piece about a growing problem for AI agents: websites are starting to block them. Amazon blocking Meta's Muse is the obvious example. At almost exactly the same time,
Simon Willison released the **“llm-openai-decisions”** plugin (version 0.1a0) on October 7 , 2026. The lightweight library lets developers send prompts directly to OpenAI’s new Jev‑clone decision mode
arXiv:2610.08414v1 Announce Type: new Abstract: Image Bitstream Fine-grained Understanding (IBFU) aims to directly perform fine-grained classification and semantic description generation from encoded
In capital markets, operational success is built on predictable performance and low latency, where even tiny variations in packet timing alter queue priority and trade execution. At the same time, bot
arXiv:2610.07646v1 Announce Type: new Abstract: Vision Language Models (VLMs) excel on visual benchmarks but fail systematically on tasks requiring abstract reasoning. Existing benchmarks document thi
Today we're announcing OpenDocRouter: every model for document parsing under one API. There are ~4,000 OCR models on Hugging Face, and the frontier labs ship a new one nearly every month. Whatever's b
arXiv:2610.07365v1 Announce Type: new Abstract: As LLMs increasingly assist scientific writing and peer review, detecting who wrote the text is no longer sufficient: we need to determine who contribut
We released Aplomb 1 today, a 5.3B decision model with open weights. It reads up to 1M tokens of text, JSON, images, video and audio in a single request, and on our API it makes a decision on a 1M-tok
Artificial Analysis: Artificial Analysis says Mistral Large 4 is the most intelligent model from outside the US and China, achieving results comparable to DeepSeek V4.1 Flash (max) — Mistral has relea
arXiv:2610.04137v1 Announce Type: new Abstract: Agent harnesses specify the roles, instructions, tools, and communication structure used to solve a task, and the right harness depends on the query. Be
arXiv:2610.05690v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for literature search and synthesis. However, it is unclear whether they retrieve accurate bibliogr
Introducing InferBench: A benchmark testing how well frontier LLMs infer a user's priorities from their instructions. We tested 12 LLMs across 20 scenarios with 2.8k conversations to see which models
My comment on Mistral Large 4 — Hacker News. wren6991: The benchmark is saturated. Frontier models are tested with an armadillo in fishnet tights jaywalking on Mars. OK well I couldn't resist this one
Enterprises and individual developers frequently run multiple AI inference models. The right architecture can simplify how the models are called while also providing centralized governance. In this po
arXiv:2610.06852v1 Announce Type: new Abstract: Pipeline figures in ML papers must be repurposed across many canvases, including paper columns, 16:9 slides, portrait posters, 1:1 social teasers, 9:16
arXiv:2610.04911v1 Announce Type: cross Abstract: Existing deep research agents are designed primarily for text- and image-based web sources, while video reasoning systems typically assume that releva
arXiv:2606.17454v3 Announce Type: replace Abstract: AI agent performance is not just a modeling problem, it is fundamentally a systems problem. The advanced capabilities of models are realized through
arXiv:2610.03639v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to support legal practice, education, and research, yet their reliability in national legal systems o
arXiv:2610.02878v1 Announce Type: new Abstract: We evaluate the performance of Vision Language Models in Visual Question Answering (VQA) when questions violate Grice's maxims. To do this, we use VLMs
arXiv:2603.29112v2 Announce Type: replace Abstract: We introduce GISTBench, a benchmark for evaluating Large Language Models' (LLMs) ability to understand users from their interaction histories in rec
We turned Claude Code, Codex, Hermes, Pi, @opencode and other coding harnesses into RL environments. No changes to the harnesses, no changes to the training code. Any open model, any task set, fully o
Banger paper from Google. It's standard practice to sample several agent rollouts and trust the answers they agree on. This Google paper shows that agreement can hide shared errors, while disagreement
Whenever I feed codebases into local models (Qwen, DeepSeek R1) or API models, the biggest annoyance is prompt pollution: - Lockfiles (`Cargo.lock`, `package-lock.json`) burning 30,000+ tokens for zer
Help with this list. Give me your recommendation on 'not supported anymore' GPUs. Looking for budget and Vulkan friendly options. Most of the GPU are not supported by latest CUDA / ROCm. Often with so
I just made the most expressive real-time motion harness for the Reachy Mini. It was created by: > synthesizing a massive motion dataset using Astra (over 10000 samples) > fine-tuning Qwen 3.5 4B usin
I've been using GPT-6 Astra to create 70+ page documents from 100,000+ characters of source material, and it's jump in processing is great. Here's the weird part, when I give it a massive task it work
arXiv:2610.00416v1 Announce Type: new Abstract: Evaluating large language models becomes increasingly challenging as their capabilities advance: benchmarks can saturate, public test sets risk contamin
arXiv:2610.00316v1 Announce Type: cross Abstract: Voice agents enable low-latency, natural interaction, yet their ability to faithfully ground responses in external documents remains underexplored. We
arXiv:2610.01023v1 Announce Type: cross Abstract: Coding agents can return plausible patches that omit required behavior. These failures are hard to review because long traces and confident summaries
Hi all! I am a former SWE who has recently transitioned into inference engineering. I launched a side hustle a few months back and I've just taken it full time due to excessive demand. Its been such a
More agents don't mean higher performance. There is a coordination bottleneck to consider. Not to mention the unnecessary costs. So how many of a multi-agent team's actions actually help it finish the
arXiv:2502.14994v2 Announce Type: replace Abstract: The rapid advancement of AI-generated video poses challenges to digital authenticity and security. Current detection methods, often trained on speci
arXiv:2610.00084v1 Announce Type: new Abstract: Detailed profession-specific system prompts raise token use and estimated cost per response without a consistent accuracy gain. We evaluate Scientific A
The same model, with the same weights, scores 62% in one agent harness and 33% in another. @adithya_s_k and the @huggingface team just released the ultimate guide to multi-harness RL, and it's one of
arXiv:2610.00972v1 Announce Type: new Abstract: As LLM agents undertake increasingly complex, long-horizon tasks, verifying their outputs becomes increasingly challenging. We study how verification ca
arXiv:2610.00994v1 Announce Type: new Abstract: Existing synthetic image evaluators typically provide only a scalar quality score and do not identify the image regions that support it. We introduce VI
arXiv:2609.39238v1 Announce Type: new Abstract: An agent that moves must recognise a place from a viewpoint it has never seen. We introduce 4MT-VLM, a dataset of procedurally generated landscapes, eac
arXiv:2603.28005v2 Announce Type: replace Abstract: When an LLM judge only has to assign a three-way support label to a candidate answer given a reference, does asking it to decompose the answer into
arXiv:2609.40324v1 Announce Type: new Abstract: We present Cogentic, a multi-agent harness for automated proof discovery on open research problems. While frontier language models can generate strong m
arXiv:2609.39378v1 Announce Type: new Abstract: Real-world embodied tasks, from everyday activities to professional procedures, require agents to act under physical constraints while tracking evolving
arXiv:2609.40340v1 Announce Type: new Abstract: Evolutionary search with large language models (LLMs) can stall when progress requires external knowledge the model lacks. Supplying relevant documents
arXiv:2609.38512v1 Announce Type: new Abstract: Voice agents in production must handle several requests, background speech, and customers who lose patience. We introduce VAmoS Energy, a benchmark that
Hybrid search, reranking, query decomposition, and query expansion are often treated as must-haves for good RAG. We wanted to see how much each actually helped, so we tested them. Same model, same emb
arXiv:2609.36368v1 Announce Type: cross Abstract: Large foundation models have been introduced with the promise of efficient adaptation to downstream tasks. Yet, under limited supervision, MLLMs, an i
arXiv:2609.38142v1 Announce Type: new Abstract: A small trainable advisor can steer a frozen language-model executor using natural-language advice. In addition to learning from task rewards, the advis
arXiv:2609.37310v1 Announce Type: cross Abstract: With LLM watermarking being deployed commercially and now required by regulations, improving its reliability and effectiveness has become crucial. Yet
arXiv:2609.36777v1 Announce Type: new Abstract: Frontier coding agents can now write and execute code that authors 3D environments, but whether they reliably understand 3D structure and precisely cont
arXiv:2609.36746v1 Announce Type: new Abstract: Agent skills provide a lightweight mechanism for self-evolving agents to accumulate reusable procedural knowledge without updating model parameters. How
arXiv:2609.37576v1 Announce Type: new Abstract: With the rapid advancement of text-to-image (T2I) generation, robust evaluation becomes critical yet challenging, as traditional metrics fail to capture
I built Image Studio, a local web app for Qwen-Image 2.1 on Apple Silicon. I had the model running on my Mac, but the only way to use it was the ComfyUI graph, and chaining edits meant wiring loaders,
arXiv:2609.37497v1 Announce Type: new Abstract: Modern transformer models excel at capturing semantic relationships through sentence embeddings, yet their ability to perform pragmatic reasoning remain
Spanner Omni, the deploy-anywhere version of Spanner, is now generally available, ready to power your most demanding production workloads in your on-premises data center or on other clouds. With Spann
arXiv:2512.03284v2 Announce Type: replace Abstract: Spatial reasoning in large-scale 3D environments remains challenging for current vision--language models, which are typically constrained to room-sc
arXiv:2609.38119v1 Announce Type: new Abstract: Long-form video understanding requires multimodal agents to iteratively gather evidence over many reasoning steps. However, most existing agentic method
Written by: Robin Grunewald, Supriya Mazumdar, Kelli Vanderlee Introduction Google Threat Intelligence Group (GTIG) examines vulnerability disclosure and exploitation statistics to evaluate the impact
arXiv:2609.33893v1 Announce Type: new Abstract: Large audio-language models (LALMs) produce fluent responses about audio but often hallucinate by making plausible yet ungrounded claims. Existing audio
arXiv:2609.31714v1 Announce Type: new Abstract: Building omni-modal models with physical intelligence requires fine-grained supervision that captures physical evidence such as contact, support, deform
arXiv:2609.33833v1 Announce Type: new Abstract: Adversarial attacks have long posed a fundamental threat to machine learning systems. As multimodal large language models (MLLMs) rapidly evolve and bec