Knowledge catalogue
I ran DeepSeek V4 Flash through Claude Code, OpenCode and Pi on my own benchmark, and the quality came out basically the same across all three while the time and tokens spent was wildly different. Cla
arXiv:2607.21274v1 Announce Type: cross Abstract: We present CUP, a Greek book retrieval benchmark consisting of 868 catalog records and 104 expert-annotated queries with graded relevance judgments. W
Announcing Fugu-Ultra v1.1 🐡 We’ve been thrilled by the reception to the Fugu model family. Thanks to everyone who tried it, shared feedback, and trusted Fugu with real work. Today, we’re releasing Fu
arXiv:2607.20764v1 Announce Type: new Abstract: We introduce ARBIGRAPH, a benchmark generator for evaluating whether tool-assisted language agents can retain, update, compose, and discard task-relevan
arXiv:2506.19445v4 Announce Type: cross Abstract: We introduce the largest real-world image deblurring dataset constructed from smartphone slow-motion videos. Using 240 frames captured over one second
arXiv:2607.21069v1 Announce Type: new Abstract: The original ALPHA benchmark introduced a taxonomy-aware penalty for evaluating CWE-level vulnerability prediction in Python and proposed that the penal
arXiv:2607.20462v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into clinical workflows, stressing the need for reliable traceability of model-generated output
arXiv:2607.21061v1 Announce Type: new Abstract: Affective Image Content Analysis (AICA) aims to recognize and understand emotions elicited by visual content, representing an indispensable step toward
Model quantization has become essential for efficient large language model deployment, yet existing approaches involve clear trade-offs: methods such as GPTQ and AWQ achieve practical compression but
arXiv:2607.21445v1 Announce Type: new Abstract: Quiz rooms, trivia nights, and quiz shows challenge human knowledge across a wide range of topics, from canonical facts to everyday culture. In this pap
Just saw this on huggingface: https://huggingface.co/ProCreations/grug-27b The benchmarks claim that it's quite a bit better than qwen3.6 27B original and that they reduced the amount of necessary tok
arXiv:2607.19716v1 Announce Type: new Abstract: Pain is a complex and pervasive phenomenon affecting a large percentage of the population, and accurate assessment is essential for effective clinical m
AI9Stars has released G9v3-3B an open weights language model designed to deliver strong reasoning capabilities within a lightweight 3 billion parameter size. It is released under the Apache 2.0 licens
arXiv:2607.20345v1 Announce Type: cross Abstract: Closing the gap between benchmark performance and reliable real-world operation remains a central challenge for Vision-Language-Action (VLA) humanoid
arXiv:2506.10890v2 Announce Type: replace Abstract: Graphic design plays a crucial role in both commercial and personal contexts, yet creating high-quality, editable, and aesthetically pleasing graphi
Many still debate open vs closed, and compare cost per token (accounting metric). Better to shift attention to cost per successful task. @seldo and the team at @arizeai did so across 2,400 runs. Concl
arXiv:2607.19235v1 Announce Type: cross Abstract: Theory of Mind (ToM), the ability to infer other's beliefs, intentions, and states of knowledge, is central to social interaction, yet remains challen
arXiv:2607.19356v1 Announce Type: new Abstract: Tool-using LLM agents increasingly execute high-impact actions, making runtime safety monitoring essential. We present NEXUS (Neural EXecution Utility a
arXiv:2607.18504v1 Announce Type: cross Abstract: Benchmarks for Geospatial Foundation Models (GFMs) increasingly rank models by aggregate score, but such rankings obscure why models differ: how much
arXiv:2607.20062v1 Announce Type: new Abstract: We present Solar Open 2, a 250B-A15B Mixture-of-Experts language model built for long-horizon agentic tasks, scaled up from Solar Open 1 (Solar Open 100
Hey HN, Henry & Roman here from Cactus. A small, on-device model is fast and private, but sometimes wrong, but frontier models are getting expensive pretty fast. So, we post-trained Gemma 4 E2B post-t
Holy shit wow We're partnering with @huggingface to investigate an unprecedented security incident. Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation. Shari
I wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face to steal the answers to the benchmark https://simonwillison
LWiAI Podcast #248 (June 12, 2026) reviews major AI developments, noting Anthropic’s release of Claude Fable 5, a safeguarded variant of Mythos 5, which shows benchmark improvements but raises concern
We're partnering with @huggingface to investigate an unprecedented security incident. Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation. Sharing preliminary
arXiv:2607.13737v1 Announce Type: new Abstract: For low-data and resource-constrained regimes typical of quantum chemistry, parameter-efficient learning is a key objective. Here, we propose a topology
arXiv:2607.13408v1 Announce Type: cross Abstract: Recent text-to-audio models generate high-quality audio, but often fail to follow instructions involving multiple sound events and temporal order. Thi
arXiv:2607.13413v1 Announce Type: cross Abstract: This study presents an empirical benchmarking comparison between Kolmogorov-Arnold Networks (KANs) and Multi-Layer Perceptrons (MLPs) on structured ta
arXiv:2607.13039v1 Announce Type: cross Abstract: Safety evaluations for dual-use biology assistants often measure base-model capability, refusal behavior, or jailbreak success. These metrics miss a d
arXiv:2607.13049v1 Announce Type: new Abstract: Foundation models have given robots a sophisticated brain for complex decision-making, yet deploying that intelligence into a physical platform still de
arXiv:2607.13618v1 Announce Type: new Abstract: LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the fin
arXiv:2607.11245v2 Announce Type: replace-cross Abstract: To reduce the substantial engineering effort required to test the corresponding applications from Android to OpenHarmony, migrating existing G
arXiv:2607.12818v1 Announce Type: new Abstract: Visual place recognition (VPR) is a key enabler of accurate localization and long-term autonomous navigation in robotics applications, such as loop clos
arXiv:2512.17788v2 Announce Type: replace Abstract: Multi-instance partial-label learning (MIPL) is a weakly supervised framework that extends the principles of multi-instance learning (MIL) and parti
arXiv:2607.12916v1 Announce Type: new Abstract: In this work, we introduce CoCo, a loss function aimed at learning normalized and well-structured representations. The proposed loss encourages intra-cl
Efficiency defined as score over active parameters. Removed all the models that were not on the pareto frontier. Yes I'm aware that artificialanalysis.ai aggregate benchmark isn't perfect, but I have
arXiv:2607.12602v1 Announce Type: new Abstract: Automatic voxel-level grounding of free-text findings in 3D chest Computed Tomography (CT) is critical for clinical interpretability. However, this task
arXiv:2509.21842v2 Announce Type: replace Abstract: Travel planning (TP) agent has recently worked as an emerging building block to interact with external tools/resources for travel itinerary generati
arXiv:2607.12787v1 Announce Type: new Abstract: Recent advances in multimodal large language models (MLLMs) have significantly improved the performance of multimodal emotion recognition (MER) and enab
arXiv:2607.13007v1 Announce Type: new Abstract: Simulation-based algorithms are especially suited for high-uncertainty environments such as adversarial board games with significant elements of randomn
arXiv:2602.15892v2 Announce Type: replace-cross Abstract: Visual perspective taking--inferring how the world appears from another's viewpoint--is foundational to social cognition. We introduce FlipSet
arXiv:2607.12527v1 Announce Type: new Abstract: Musculoskeletal diseases are among the leading causes of disability worldwide and create the greatest global need for rehabilitation. Because recovery,
arXiv:2607.11951v1 Announce Type: new Abstract: Large language models can write SQL, but enterprise deployment demands more than plausible text: outputs must be syntactically valid, must respect per-r
arXiv:2509.24863v2 Announce Type: replace Abstract: Deep neural networks surpass humans on many vision benchmarks, yet remain far less robust to distribution shifts such as illumination and weather ch
arXiv:2607.12067v1 Announce Type: new Abstract: Institutional equity holdings disclosed in SEC Form 13F filings provide a rich temporal record of portfolio decisions by large investment managers. Howe
arXiv:2607.12625v1 Announce Type: new Abstract: OpenClaw has emerged as a leading agent framework for complex task automation, yet it faces insufficient cross-platform GUI interaction support and a we
arXiv:2607.12771v1 Announce Type: cross Abstract: Reaction mechanisms consist of the step-by-step sequences of elementary reactions that explain chemical transformations. Learning the mechanism logic
arXiv:2607.12696v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) models have become an important approach for scaling Large Language Models (LLMs), but their inference efficiency depe
arXiv:2607.11977v1 Announce Type: new Abstract: In 2019, OpenAI released two million GPT-2 outputs-ungrammatical, half broken-to aid the detection of machine-generated text. The alignment that produce
arXiv:2607.12985v1 Announce Type: new Abstract: Aligned language models routinely misreport under non-evidential incentive pressure: they agree with a confident user or overstate certainty even when t
arXiv:2503.06796v2 Announce Type: replace Abstract: Robot design is a complex and time-consuming process that requires specialized expertise. Gaining a deeper understanding of robot design data can en
arXiv:2506.12339v2 Announce Type: replace-cross Abstract: We present SheetMind, a modular multi-agent framework powered by large language models (LLMs) for spreadsheet automation via natural language
arXiv:2607.13028v1 Announce Type: cross Abstract: Training robust autonomous driving agents requires a simulator that is fast enough for reinforcement learning at scale, realistic enough to ground beh
arXiv:2607.12545v1 Announce Type: cross Abstract: Adversarial robustness research has produced hundreds of defended models over the past decade, yet the literature almost universally reports robustnes
arXiv:2607.12441v1 Announce Type: new Abstract: Wikipedia plays a key role in shaping public understanding of science, and its openly accessible revision history is a unique record of how scientific k
The NVIDIA Nemotron Model Reasoning Challenge on Kaggle attracted over 5,000 participants who all began from the same open model, benchmark, and infrastructure. The strongest solutions treated reasoni
In this post, we introduce the UI for optimized generative AI inference recommendations in Amazon SageMaker AI Studio, a low-code no-code (LCNC) experience. The API already gives you programmatic acce
We trained and released DSpark speculators for Kimi-K2.6 and Kimi-K2.7-Code on @huggingface, with native serving support in @vllm_project. Across six benchmarks in our batch-size-1 evaluation: Kimi-K2
Grok places second after Fable on real-world software engineering Grok 4.5 from @SpaceXAI places #2 on the APEX-SWE leaderboard at 51.2% Pass@1 (±6.0), behind Fable 5 (65.5% ±6.2) on our benchmark for
arXiv:2602.05459v2 Announce Type: replace Abstract: Offline goal-conditioned reinforcement learning (GCRL) is typically benchmarked by the best tuned success rate of each method. This score measures a