Search: “models”

This post announces an 8-9 billion parameter language model optimized for efficiency and speed, trained on a very large token dataset, and positioned as accessible for users with limited GPU resources

arXiv:2602.02898v2 Announce Type: replace Abstract: Language model benchmarks are pervasive and computationally-efficient proxies for real-world performance. However, many recent works find that bench

arXiv:2605.28020v1 Announce Type: new Abstract: With the rapid progress of large language models (LLMs), reliably evaluating the capabilities of pre-trained LLMs has become increasingly important. The

Today we're announcing ESMFold2, an open scientific engine to power prediction, design, and discovery across protein biology. The new model delivers state of the art performance on protein interaction

arXiv:2605.23966v1 Announce Type: cross Abstract: Optimization modeling serves as the pivotal bridge between natural-language problem descriptions and optimization solvers, and remains a cornerstone f

Tesla's Model S, designed by Chief Designer Franz von Holzhausen, marked a significant milestone spanning from its first customer deliveries in 2012 to its final production run in 2026, representing 1

arXiv:2508.20614v3 Announce Type: replace-cross Abstract: Amortized Bayesian model comparison (BMC) enables fast probabilistic ranking of models via simulation-based training of neural surrogates. How

arXiv:2605.09154v1 Announce Type: new Abstract: We introduce a predictive model that estimates the pre-training loss of large models from model size (N), batch size (B) and number of weight updates (K

arXiv:2605.05615v2 Announce Type: replace Abstract: Large language models (LLMs) impose rapidly growing energy demands, creating an emerging energy and carbon crisis driven by large-scale inference. S

Post-training quantization is a technique that reduces model size and improves inference performance by converting weights and activations to lower precision formats after training is complete. NVIDIA

The Harness is a Context Manager on Behalf of the Model What happens when the context window fills up and who decides? This decision is external to the model - The Harness designer must have some opin

arXiv:2604.27311v1 Announce Type: cross Abstract: The advent of Large Language Models (LLMs) has significantly transformed tasks across Software Engineering. In the context of Business Process Managem

arXiv:2502.02779v3 Announce Type: replace-cross Abstract: Head computed tomography (CT) imaging is a widely-used imaging modality with multitudes of medical indications, particularly in assessing path

arXiv:2510.04686v2 Announce Type: replace-cross Abstract: Model merging combines independent solutions with different capabilities into a single one while maintaining the same inference cost. Two popu

Ethan Mollick comments on an impressive AI model achievement, noting that the performance gap between models appears consistent with historical trends based on his assessment.

arXiv:2601.21257v2 Announce Type: replace Abstract: Advancing beyond single monolithic language models (LMs), recent research increasingly recognizes the importance of model collaboration, where multi

arXiv:2604.18327v1 Announce Type: cross Abstract: Reward models (RMs) are central to aligning large language models (LLMs) with human preferences, powering RLHF and advanced decoding strategies. While

Claude Token Counter, now with model comparisons I upgraded my Claude Token Counter tool to add the ability to run the same count against different models in order to compare them. As far as I can tel

dear god lol - new kimi model is a fucking beast. GPT 5.4 level coding, 76% cheaper than opus 4.7 and 100% open source / free to use, i mean look at this: > kimi k2.6 can code continuously for 12 hour

arXiv:2510.09065v2 Announce Type: replace-cross Abstract: We introduce MMAudioSep, a generative model for video/text-queried sound separation that is founded on a pretrained video-to-audio model. By l

arXiv:2604.14769v1 Announce Type: new Abstract: The pre-training and fine-tuning paradigm has become the dominant approach for model adaptation. However, conventional pre-training typically yields mod

arXiv:2508.20705v3 Announce Type: replace Abstract: Recent advances in self-supervised learning for EEG representation have largely relied on masked reconstruction, where models are trained to recover

Last week, Anthropic announced Project Glasswing alongside Claude Mythos Preview, a model they described as so powerful at finding vulnerabilities they couldn't release it. The announcement featured A

arXiv:2509.26404v2 Announce Type: replace-cross Abstract: Fingerprinting Large Language Models (LLMs)is essential for provenance verification and model attribution. Existing fingerprinting methods are

arXiv:2604.09378v1 Announce Type: cross Abstract: Agent ecosystems increasingly rely on installable skills to extend functionality, and some skills bundle learned model artifacts as part of their exec

Open everything 🔥 Open Harness, Model Choice, Open Memory (take it wherever you need), Open Protocols basically we’re in the middle of a model war, they all make great models but the optimal arrangeme

Amazon released the Nova 2 model family on December 2, 2025 at AWS re:Invent, comprising Nova 2 Lite, Nova 2 Pro (Preview), Nova 2 Omni, and Nova 2 Sonic, all featuring a 1-million-token context wi...

arXiv:2507.22612v3 Announce Type: replace-cross Abstract: Speech-to-text alignment is a critical component of neural text to speech (TTS) models. Autoregressive TTS models typically use an attention m

arXiv:2610.22343v1 Announce Type: cross Abstract: Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometr

arXiv:2610.20982v1 Announce Type: cross Abstract: We report Monte Carlo simulations of a continuous spin variant of the Blume-Capel model. We compute the phase diagrams in the crystal-field-temperatur

arXiv:2609.31784v2 Announce Type: replace Abstract: Open-weight models are often released, fine-tuned, aligned, merged, and re-released, making provenance audits ask not only whether checkpoints are r

arXiv:2610.18378v1 Announce Type: new Abstract: Applying process-based Earth science models to new sites and questions often requires expert calibration decisions. Reliable delegation to language-mode

arXiv:2610.18430v1 Announce Type: cross Abstract: Federated fine-tuning of large language models (LLMs) allows multiple clients to collaboratively adapt models without sharing raw data, offering stron

arXiv:2402.03694v3 Announce Type: replace-cross Abstract: Network traffic analysis in cloud and internet service provider (ISP) networks increasingly relies on machine learning (ML) models, as encrypt

Microsoft AI: Microsoft unveils MAI-Base-2, a MoE model with 2.8T total and 80B active parameters, claiming 4x-12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base — In June, we announced M

arXiv:2610.15883v1 Announce Type: cross Abstract: Text wrapped in a trigger can be made behaviorally invisible to a language model: omitted from summaries and quotations, ignored in question answering

arXiv:2610.15625v1 Announce Type: new Abstract: World Action Models (WAMs) unify visual world modeling and action generation, yet their real-world deployment is heavily bottlenecked by iterative actio

Just saw this demo from OpenBMB for MiniCPM-V 4.7, and the on-device capability is pretty impressive. A 35B-A3B MoE vision-language model running locally on mobile devices is already interesting, but

arXiv:2610.14913v1 Announce Type: cross Abstract: Reinforcement learning (RL) is essential for adaptive decision-making, e.g., robust Meta-RL and domain randomization (DR), and efficient post-training

arXiv:2610.14762v1 Announce Type: new Abstract: Model merging combines fine-tuned experts into one network without retraining, and closed-form methods compute each merged layer from statistics of its

arXiv:2610.16166v1 Announce Type: new Abstract: Interactive world models must capture how multiple participants jointly change a shared environment while observing different parts of it. Competitive g

arXiv:2610.14558v1 Announce Type: new Abstract: Large pretrained models have achieved remarkable success across diverse domains, yet adapting them to multiple downstream tasks remains challenging: ful

DISCLAIMER: the post and the model card was made with the assist of AI. Hiya, I’m releasing Recursive BitNet N-Gram 102M, a small experiment combining ternary weights, shared transformer layers, and h

arXiv:2512.14014v2 Announce Type: replace Abstract: World models have shown great utility in improving the task performance of embodied agents. While prior work largely focuses on pixel-space world mo

I've been building ANCHOR, a web automation agent that runs entirely on Ollama. You describe a task in plain language ('Register Maria Silva, IT department, contractor, and save'), point it at a site,

arXiv:2602.02639v2 Announce Type: replace-cross Abstract: LLM self-explanations are often presented as a promising tool for AI oversight, yet their faithfulness to the model's true reasoning process i