Search: “models”

A study from @Stanford showed that 71.3% of chatgpt queries could be accurately answered by a local model. I suspect a major part of enterpr…

A study from @Stanford showed that 71.3% of chatgpt queries could be accurately answered by a local model. I suspect a major part of enterprise AI workloads could be run locally too for free (compared

local-aiclem-delangue--x

Research

A Theoretical Interpretation of In-Context Learning via Probabilistic Modeling

DGX agent

arXiv:2606.28926v1 Announce Type: cross Abstract: In-context learning (ICL) is an emerging paradigm that employs the semantic information inherent in large language models (LLMs) for generating answer

researcharxiv-cs-lg

30 Jun 2026

Arko-T: A Foundation Model for Text-to-Structured 3D Generation

model-releasesarxiv-cs-lg

arXiv:2606.30429v1 Announce Type: new Abstract: Text-to-3D systems can now synthesize a mechanical part from a single sentence, yet the result is a shape to render, not a design to edit. We present Ar

As open models get stronger, more workloads move into the competitive inference market. That pushes the real fight toward speed, cost, relia…

applicationstogether-ai--x

As open models get stronger, more workloads move into the competitive inference market. That pushes the real fight toward speed, cost, reliability, and control. Together AI is where open models become

Safety

Can MLLMs Critique Like Humans? Evaluating Open-Ended Aesthetic Reasoning in Multimodal Large Language Models

arXiv:2606.29689v1 Announce Type: new Abstract: Open-ended aesthetic critique is a challenge for multimodal large language models (MLLMs): unlike multiple-choice aesthetic benchmarks, it has no single

safetyarxiv-cs-cl

ClusterStyle: Modeling Intra-Style Diversity with Prototypical Clustering for Stylized Motion Generation

arXiv:2512.02453v2 Announce Type: replace Abstract: Existing stylized motion generation models have shown their remarkable ability to understand specific style information from the style motion, and i

local-aiarxiv-cs-cv

Cognitive World Models for Process-Level Social Influence Evaluation

arXiv:2606.29495v1 Announce Type: new Abstract: Social influence dialogue changes user behavior by altering internal cognitive states. The central evaluation question is whether the user's beliefs, de

Event-Conditioned Diagnostics of Kinematic, Contact, and Object-Permanence Fields in Passive Object-State World Models

arXiv:2606.28455v1 Announce Type: cross Abstract: World models can predict future physical states, but prediction accuracy alone does not explain how physical information is organized and used inside

tutorialsarxiv-cs-ai

Event-VLA: Action-Conditioned Event Fusion for Robust Vision-Language-Action Model

arXiv:2606.29384v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have become an important paradigm of embodied AI. However, existing VLA models typically assume well-lit and stable

applicationsarxiv-cs-cv

Exploiting Vision Encoder Vulnerabilities for Universal Adversarial Perturbations on Large Vision-Language Models

arXiv:2412.08108v3 Announce Type: replace-cross Abstract: Large Vision-Language Models (LVLMs) have achieved remarkable performance on multimodal tasks but remain highly vulnerable to small adversaria

researcharxiv-cs-cl

Tools

Featuring Every Eval Ever Results on Hugging Face Model Pages

Hugging Face introduced a feature that displays evaluation results from the 'Every Eval Ever' (EEE) community evaluation initiative directly on model pages, providing users with comprehensive benchmar

toolshugging-face

Field Order Should Not Matter: Permutation-Invariant Embedding Model Fine-Tuning for Structured Metadata Retrieval

arXiv:2606.30473v1 Announce Type: cross Abstract: We study retrieval over catalogs of structured metadata, where each record is a small schema whose fields answer different kinds of query. Embedding a

Fine-tune Amazon Nova models for accurate email data extraction

In this post, you'll learn how fine-tuning Amazon Nova models using Amazon SageMaker AI addresses these specific issues by teaching the models to recognize your exact data patterns, distinguish betwee

tutorialsaws-ml-blog

Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding

arXiv:2603.19235v2 Announce Type: replace Abstract: While Multimodal Large Language Models demonstrate impressive semantic capabilities, they often suffer from spatial blindness, struggling with fine-

tutorialsarxiv-cs-cv

Industry

Google's new Nano Banana 2 Lite image model is its fastest and cheapest yet

Google introduced Nano Banana 2 Lite, its fastest and most cost-efficient image model built for high throughput, speed and scale. The model generates text-to-image outputs in about four seconds and co

industryars-technica

Grounding Sim-to-Real Generalization in Robotic Manipulation: An Empirical Study with Vision-Language-Action Models

arXiv:2603.22876v2 Announce Type: replace-cross Abstract: Learning a generalist control policy for robotic manipulation typically relies on large-scale datasets. Given the high cost of real-world data

Introducing TabFM: A zero-shot foundation model for tabular data

TabFM is a foundation model designed for tabular data classification and regression that eliminates the need for manual model training, hyperparameter tuning, and complex feature engineering by framin

researchgoogle-research

Model Predictive Current Control with Harmonic Correction for Single-Phase AC-DC EV Charging

arXiv:2606.30397v1 Announce Type: cross Abstract: The increasing integration of Electric Vehicles (EVs) has imposed a growing harmonic challenge on the power grid. For AC/DC Power Factor Correction (P

NIVA: A Multimodal Foundation Model for Actionable Earth System Intelligence

arXiv:2606.28546v1 Announce Type: new Abstract: Recent advances in AI-driven weather and climate modeling have improved forecast skill while reducing computational cost. However, existing data-driven

tutorialsarxiv-cs-lg

Physics-Grounded Disentangled Flow Modeling for Brain Disease Progression Trajectory

arXiv:2606.28630v1 Announce Type: new Abstract: Forecasting longitudinal brain lesion evolution is critical for disease monitoring and treatment planning. Existing approaches typically learn a direct

tutorialsarxiv-cs-cv

Surrogate Modeling for Explainable Predictive Time Series Corrections

arXiv:2412.19897v3 Announce Type: replace-cross Abstract: We introduce a local surrogate approach for explainable time-series forecasting. An initially non-interpretable predictive model to improve th

local-aiarxiv-cs-lg

The CRISTAL Method: Neurosymbolic analysis from AI-synthesized world models

arXiv:2606.29799v1 Announce Type: new Abstract: This project introduces the CRISTAL Method (Coherent Reliable Intentional Synthesis of Truthful Analysis Logic), a neurosymbolic framework for automatin

Towards Generalizable and Evidential Nuclear Magnetic Resonance-Based Molecular Structure Elucidation via Large Language Model Agent

arXiv:2606.29776v1 Announce Type: cross Abstract: Nuclear Magnetic Resonance (NMR) spectroscopy is the gold standard for molecular structure elucidation, yet interpreting complex spectra for unknown m

Safety

TrafficAlign: Aligning Large Language Models for Traffic Scenario Generation

arXiv:2606.29097v1 Announce Type: new Abstract: Recent research has investigated the use of large language models (LLMs) to generate traffic scenarios for autonomous driving. However, pretrained LLMs

safetyarxiv-cs-cv

Safety

Value-Action Alignment in Large Language Models under Privacy-Prosocial Conflict

arXiv:2601.03546v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used to simulate decision-making tasks involving personal data sharing, where privacy concerns a

safetyarxiv-cs-ai

What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs

arXiv:2606.28615v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in high-stakes domains, where free-text explanations such as chain-of-thought and post-hoc rati

tutorialsarxiv-cs-ai

Agents

CascadeOcc: Rethinking 3D Occupancy World Models with Cascaded VQ Representations

arXiv:2606.27644v1 Announce Type: new Abstract: This letter proposes CascadeOcc, a novel occupancy world model that prioritizes intrinsic structural hierarchy over extrinsic auxiliary modalities for a

agentsarxiv-cs-cv

Agents

Conventional model routing sucks. It passes benchmarks but fails to write code you'd actually merge. Introducing Devin Fusion, a new hybrid-…

Conventional model routing sucks. It passes benchmarks but fails to write code you'd actually merge. Introducing Devin Fusion, a new hybrid-model harness for agentic coding. In testing, it reduces the

agentscognition-ai--x

Developmental approach reveals the statistical learning of Neural Language Models: Transformers generalize from the most abstract statistical patterns

arXiv:2606.27460v1 Announce Type: new Abstract: In this study, we use a developmental approach to investigate the statistical learning and mental representation of neural language models (NLM). A seri

tutorialsarxiv-cs-cl

FailSafe: Reasoning and Recovery from Failures in Vision-Language-Action Models

arXiv:2510.01642v3 Announce Type: replace Abstract: Recent advances in robotic manipulation have integrated low-level robotic control into Vision-Language Models (VLMs), extending them into Vision-Lan

applicationsarxiv-cs-ro

Global Explanations for Multivariate Time Series Forecasting Models via K-Order Markov Approximations

arXiv:2606.27599v1 Announce Type: cross Abstract: While many explainable AI (XAI) methods have been proposed, most are not designed for time-series forecasting models and often rely on the implicit as

applicationsarxiv-cs-ai

Masked Language Flow Models

arXiv:2606.27617v1 Announce Type: new Abstract: Masked Diffusion Models (MDMs) promise fast, parallel language generation, but their reverse transition factorises across token positions -- an approxim

researcharxiv-cs-cl

Mitigating LLM-based p-Hacking by Preregistering for the Next LLM

arXiv:2606.27687v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to generate, classify, and annotate data whose outputs feed downstream hypothesis tests. However, L

ProactiveBench: Benchmarking Proactiveness in Multimodal Large Language Models

model-releasesarxiv-cs-cv

arXiv:2603.19466v2 Announce Type: replace Abstract: Effective collaboration begins with knowing when to ask for help. For example, when trying to identify an occluded object, a human would ask someone

Scaling limit of the Random Language Model

arXiv:2606.28105v1 Announce Type: cross Abstract: We develop a quantitative theory of the Random Language Model (RLM), an ensemble of stochastic context-free grammars, in a scaling limit where the num

researcharxiv-cs-cl

Synthesize the big picture and analyze trends with BigQuery's AI.AGG function

model-releasesgoogle-cloud-ai

We recently announced the preview of the BigQuery AI.AGG() function. With AI.AGG(), you can use natural-language instructions within a single line of SQL to summarize or synthesize information over mi

Understanding Rollout Error in Graph World Models

arXiv:2606.27780v1 Announce Type: new Abstract: World models are often used for planning by rolling learned dynamics forward. Many planning environments, however, are not vectors or images; they are g

local-aiarxiv-cs-ai

VLM-Aware Meta-Optic Front-End Design for Frozen Vision-Language Models

model-releasesarxiv-cs-cv

arXiv:2606.27646v1 Announce Type: new Abstract: Conventional machine-vision pipelines typically rely on high-quality optics that produce clean, human-interpretable images, and optical design has there

Agents

We are starting to role out our Trace Judge model to early partners today Designed to detect errors in agent trajectories at 1/100th of the …

We are starting to role out our Trace Judge model to early partners today Designed to detect errors in agent trajectories at 1/100th of the cost compared to closed models If interested in early access

agentsharrison-chase--x

Industry

You don't need Fable for the most complex tasks, from training models for protein prediction to optimising compilers Our open source Zenith …

You don't need Fable for the most complex tasks, from training models for protein prediction to optimising compilers Our open source Zenith harness takes base models to the top of FrontierSWE via adap

industryemad-mostaque--x

As this post points out, contrary to what many say, the US government could absolutely effectively ban open weights models. That doesn’t mea…

applicationsethan-mollick--x

As this post points out, contrary to what many say, the US government could absolutely effectively ban open weights models. That doesn’t mean you won’t be able to download the weights & run them, but

Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions

arXiv:2606.26982v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly being integrated into mental health support tools and other psychologically sensitive conversational app

researcharxiv-cs-ai

Estimating Uncertainty in Classifier Performance with Applications to Large Language Models and Nested Data

arXiv:2606.26422v1 Announce Type: new Abstract: Researchers increasingly use text classification--supervised models or large language models--to measure constructs from natural language, providing met

researcharxiv-cs-ai

How Good Can Linear Models Be for Time-Series Forecasting?

arXiv:2606.27282v1 Announce Type: new Abstract: Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation model

researcharxiv-cs-lg

Stochastic Gradient Optimization with Model-Assisted Sampling

model-releasesarxiv-cs-lg

arXiv:2606.27171v1 Announce Type: new Abstract: This work addresses the problem of variance in stochastic gradient estimation for machine learning optimization. Deep learning relies on mini-batch meth

CustomX: Unified Character, Action, and Scene Customization in Video World Models

arXiv:2512.17796v2 Announce Type: replace Abstract: Recent advances in world models have greatly enhanced interactive environment simulation. Existing methods mainly fall into two categories: (1) stat

researcharxiv-cs-cv

25 Jun 2026

Efficient and Trainable Language Model Test-Time Scaling via Local Branch Routing

arXiv:2606.25354v1 Announce Type: new Abstract: Test-time scaling improves language-model reasoning, but existing approaches often face a difficult trade-off: long chain-of-thought sampling remains si

local-aiarxiv-cs-cl

25 Jun 2026

FlowID : Enhancing Forensic Identification with Latent Flow-Matching Models