The Advantage of Fine-Grained Training
arXiv:2509.05130v2 Announce Type: replace Abstract: In classification problems, models are trained to predict a class label based on the input data features. However, class labels are organized hierar
Knowledge catalogue
arXiv:2509.05130v2 Announce Type: replace Abstract: In classification problems, models are trained to predict a class label based on the input data features. However, class labels are organized hierar
arXiv:2607.26838v1 Announce Type: new Abstract: In this paper we show that the generalization error of AdaBoost is Thetaig(frac{dln(ngamma^{2}/d)}{ngamma^2}+frac{ln(1/elta)}{n}ig), where gamma is the
arXiv:2607.26854v1 Announce Type: new Abstract: Accurate multi-week dengue forecasting supports timely vector-control interventions, outbreak preparedness, and healthcare resource allocation. However,
arXiv:2607.27027v1 Announce Type: new Abstract: Gradient-boosted trees dominate tabular machine learning, yet canonical correlation analysis has always relied on linear or neural encoders. We propose
arXiv:2607.26977v1 Announce Type: new Abstract: Travel planning is a demanding stress test for tool-using LLM agents: a usable itinerary is a single artifact that must be right along many axes at once
Its a custom Swift/Metal inference engine that runs Gemma 4 26B-A4B-IT on M-series Macs with very low RAM. It uses ~2GB instead of ~14 GB. The result is reportedly 5–6 tok/s on an 8 GB M2 MacBook Air
I've tried grok and Claude's deep research mode and I was amazed with the speed considering the amount of sources analysed. Is there anything as fast that can run locally? My guess would be that to ru
arXiv:2607.26795v1 Announce Type: new Abstract: Span-guided rewriting aims to preserve meaning by localizing edits to annotated harmful spans, but the same constraint can leave harmful intent insuffic
arXiv:2607.26555v1 Announce Type: new Abstract: Multimodal fake news detectors often generalize poorly across domains because they learn to trust unreliable evidence: domain-specific shortcuts amplifi
A new TIL on adding custom MCP servers to both the ChatGPT and Claude regular chat interfaces - it's a little less obvious than I had hoped, but I got there in the end https://til.simonwillison.net/ll
arXiv:2607.25668v1 Announce Type: new Abstract: Identifying cost-effective indigenous building materials that minimise heat penetration through walls is critical for indoor thermal comfort in low-inco
After deployment, we applied GPT-5.6 Sol to advance the frontier of efficiency by making itself more efficient to run. The results: - 20% lower serving costs from production GPU kernel improvements. -
arXiv:2607.24882v1 Announce Type: cross Abstract: Modern coding agents are usually evaluated by whether they eventually produce a correct patch, but patch generation depends on an earlier context-acqu
arXiv:2607.25634v1 Announce Type: new Abstract: We present AIriskEval-edu Demo, a platform that audits the pedagogical quality of instructional explanations and provides explainable audit results. The
arXiv:2607.25502v1 Announce Type: new Abstract: Recent training-time defenses against neural backdoors isolate a benign subset from poisoned training data, to learn a backdoor-free model from it. In t
add rdna3.5, and 3 to mmq configs so they can be tuned independently. (#26199) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED
RPC: add tensor_memset (#25912) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework Linux: Ubuntu
server : add trace logging for slot similarity checking (#26271) Adds trace logging in server-context.cpp for slot similarity checking during prompt cache slot selection, including skip reasons and si
vendor: update BoringSSL to 0.20260728.0 (#26241) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFrame
sycl: contiguous fast path + 32-bit index math for unary elementwise ops (#25946) sycl: contiguous fast path + 32-bit index math for unary elementwise ops sycl: use fastdiv for elementwise index math
llama: move suppress_tokens handling to common/sampling (#26276) llama: move suppress_tokens handling to common/sampling address security issues rm has_logit_bias Website: https://llama.app macOS/iOS:
arXiv:2607.25318v1 Announce Type: new Abstract: Dataset distillation aims to synthesize compact datasets that can approximate the performance of full-data training while significantly reducing computa
arXiv:2607.25600v1 Announce Type: cross Abstract: Retrieval-augmented generation improves knowledge-intensive question answering, but indiscriminate retrieval can introduce irrelevant evidence and unn
BREAKING: Grok 4.5 (high) ranks #1 on the HighWalk benchmark, which tests how well AI agents update technical specifications from code changes. Grok delivered the best combination of quality and opera
BREAKING: Grok 4.5 ranked #1 on LaurenBench with a score of 56.9%, ahead of Claude Sonnet 5, GLM 5.2, Claude Opus 5, Kimi K3 and GPT-5.6. The benchmark tests real-world AI agents across conversation,
Sorry if flair is wrong. I decided to finally get a ChatGPT subscription after some conversations with it about health issues with my dog. I've only used AI for coding work, primarily Claude, but I fe
ComfyUI Face Swap Tutorial: Fast, Clean Results VFX artist @heydoughogan walks through a fully automated face swap workflow built inside ComfyUI - combining Florence 2, SAM2, and WAN Video into a sing
arXiv:2501.17296v4 Announce Type: replace-cross Abstract: Multiphysics simulations play an essential role in accurately modeling complex interactions across diverse scientific and engineering domains
arXiv:2607.25754v1 Announce Type: new Abstract: Cooperative navigation of multi-agent UAVs in complex environments faces key challenges including local optima traps, sparse rewards, learning imbalance
arXiv:2607.25356v1 Announce Type: cross Abstract: Data quality profiling -- computing missing-value rates, duplicate fractions, outlier densities, and functional-dependency violations -- is foundation
arXiv:2607.25129v1 Announce Type: new Abstract: Medical coding is the task of assigning a set of diagnosis and procedure codes for a hospitalization using recorded notes. It requires aggregating infor
DeepSeek V4 Flash isn't just for inference anymore. Fine-tune it on Fireworks with supervised fine-tuning, preference tuning, and combined preference optimization from the managed UI. Reinforcement le
arXiv:2607.25750v1 Announce Type: new Abstract: Low-rank adaptation (LoRA) fine-tuning has made it cheap and easy to customize open-weight image generation models for specific tasks, including the pro
arXiv:2607.25959v1 Announce Type: cross Abstract: Wikipedia and Wikidata are widely used for information access, LLM pre-training, and retrieval-augmented generation. Their knowledge is deeply connect
arXiv:2607.24745v1 Announce Type: cross Abstract: Key Information Extraction (KIE) is vital for many document applications, but creating training datasets is traditionally a time-consuming manual proc
Dreaming in Voxels: How AI is Generating Playable Minecraft Worlds Generative AI has conquered images, video, text. But what about interactive 3D environments? We trained models on billions of cubes t
Saw that the Asus Ascent 1tb was going for $3,950 from a few sources, couldn't stop thinking about it, finally just went ahead and did it. Am I completely insane? Will I regret this? I can't imagine t
arXiv:2607.24981v1 Announce Type: new Abstract: Vision Transformer detectors now approach the accuracy of CNNs but remain difficult to deploy on NPUs and microcontrollers because key components, inclu
arXiv:2607.25123v1 Announce Type: new Abstract: Experience replay remains one of the most practical and useful algorithmic tools in the deep reinforcement learning (DRL) toolbox. Aside from the limite
Day one threads are the least useful thing we produce here and we produce a lot of them. Model drops, forty people run their favourite prompt, half say it's the best thing ever and half say benchmaxxe
arXiv:2607.21859v2 Announce Type: replace Abstract: Constructing causal directed acyclic graphs (DAGs) is a core step in biomedical causal analysis, yet it remains a largely manual process. Analysts m
arXiv:2607.25794v1 Announce Type: new Abstract: Fine-grained food visual--semantic understanding requires models to capture subtle distinctions across ingredients, cooking methods, doneness, color, te
I've got better results than expected for 768gb DDR5 and 2x5090. Using fork https://github.com/pwilkin/llama.cpp/tree/kimi-k3-text and https://huggingface.co/GrEarl/Kimi-K3-GGUF Q2_K quant. Prefill sp
arXiv:2607.25322v1 Announce Type: new Abstract: Multimodal drug discovery enables drug representation learning beyond chemical structure by incorporating cellular responses such as gene expression and
Gary Marcus has argued that both OpenAI and Anthropic must increase their pricing, citing the rapid decline in token costs and competition from open‑source models. He presents a binary choice: either
arXiv:2607.24788v1 Announce Type: new Abstract: As Large Language Models scale to increasingly long contexts, the memory I/O and computational overhead of the Key-Value (KV) cache during decoding emer
GPT-5.6 Sol has been used to solve open problems in mathematics. So why was it struggling with ARC-AGI-3, a benchmark of 2D puzzle games? We investigated. The harness was not letting it remember what
arXiv:2607.24835v1 Announce Type: new Abstract: Weakly supervised hierarchical models exhibit a persistent asymmetry: coarse lesion-type features are preserved under reconstruction while fine-grained
arXiv:2508.05899v3 Announce Type: replace Abstract: 3D scene generation plays a crucial role in gaming, artistic creation, virtual reality, and many other domains. However, current 3D scene design sti
After deployment, we applied GPT-5.6 Sol to advance the frontier of efficiency by making itself more efficient to run. The results: - 20% lower serving costs from production GPU kernel improvements. -
arXiv:2607.24861v1 Announce Type: cross Abstract: Question answering (QA) over complex documents requires models to retrieve and integrate evidence distributed across distant document regions and moda
arXiv:2607.25297v1 Announce Type: cross Abstract: The rapid development of large language model (LLM) agents has enabled their broad adoption across diverse real-world tasks. To standardize interactio
OpenAI has released the open‑source Codex Security CLI, which can be installed with `npm install @OpenAI/codex-security` or run directly via `npx @OpenAI/codex-security@latest --help`. The tool scans
arXiv:2607.25904v1 Announce Type: new Abstract: Graphical user interface task evaluation aims to determine whether a GUI agent has successfully completed a user instruction. Automated GUI task evaluat
arXiv:2603.14558v3 Announce Type: replace Abstract: Recruiters and job seekers rely on search systems to navigate labor markets, making candidate matching engines critical for hiring outcomes. Most sy
arXiv:2607.25626v1 Announce Type: new Abstract: Decoding speech information directly from scalp electroencephalography (EEG) into text provides a potential non-invasive neural communication pathway fo
arXiv:2607.25125v1 Announce Type: new Abstract: Despite rapid progress in Multi-modal Large Language Models (MLLMs), understanding long-form videos is still bottlenecked by limited context windows. Wh
arXiv:2607.25748v1 Announce Type: new Abstract: Objective: Concept bottleneck models route prediction through interpretable intermediate variables, and their validity is normally judged by how accurat
Microsoft is working on an AI 'super app' that combines Copilot's chat, coding, and agentic capabilities. During an earnings call on Wednesday, Microsoft CEO Satya Nadella said the app will span 'both
arXiv:2607.25538v1 Announce Type: new Abstract: Swarm Learning is a decentralized collaborative learning mechanism that allows multiple organizations to train a shared model without central coordinati