Search: “models”

Multimodal Graph RAG for Long-range Visually Rich Document Understanding

arXiv:2606.28780v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) are widely applied to visual document understanding. However, comprehending long documents remains an issue b

Multimodal Mathematical Reasoning with Diverse Solving Perspective

DGX agent

arXiv:2507.02804v2 Announce Type: replace Abstract: Recent progress in large-scale reinforcement learning (RL) has notably enhanced the reasoning capabilities of large language models (LLMs), especial

model-releasesarxiv-cs-cl

30 Jun 2026

Nano Banana 2 Lite

model-releasessimon-willison

Nano Banana 2 Lite Also known as Gemini 3.1 Flash Lite Image (gemini-3.1-flash-lite-image in their API), this is the 'fastest and cheapest Gemini image model, engineered for velocity and scale'. I use

Not-quite-human tastes: the stylized omnivorousness of LLM survey surrogates

arXiv:2606.30085v1 Announce Type: new Abstract: Large-language models have proven to be remarkable if inconsistent parrots of public attitudes and opinions. The extent to which LLMs are able to produc

OP3DSG: Open-Vocabulary Part-Aware 3D Scene Graph Generation for Real-World Environments

arXiv:2606.29786v1 Announce Type: new Abstract: 3D scene graphs (3DSGs) provide a compact and structured abstraction of 3D environments. Although advances in foundation models have enabled open-vocabu

Research

Perspectives on Latent Factor Indeterminacy and its Implications for Data Representation

arXiv:2606.28854v1 Announce Type: cross Abstract: The common factor analytic model is related to Helmholtz and Boltzmann machines, can be conceived as a linear autoencoder, or can be thought of as a s

researcharxiv-cs-ai

REAR: Test-time Preference Realignment through Reward Decomposition

arXiv:2606.30339v1 Announce Type: new Abstract: Aligning large language models (LLMs) with diverse user preferences is a critical yet challenging task. While post-training methods can adapt models to

safetyarxiv-cs-cl

REPAIR-Bench: A Benchmark for Robot Error Perception And Interaction Recovery

model-releasesarxiv-cs-ro

arXiv:2606.29937v1 Announce Type: new Abstract: Understanding how users perceive and respond to robot failures is essential for building robust and trustworthy robot systems. Prior work, however, (i)

Reported Confidence in LLMs Tracks Commitment More Than Correctness

arXiv:2606.29490v1 Announce Type: cross Abstract: Confidence is an estimate of the probability that a chosen answer is correct. Verbal confidence reports are widely used as uncertainty measures in lar

Reward-Free Code Alignment from Pretrained or Fine-Tuned LLM: Unpacking the Trade-offs for Code Generation

arXiv:2606.28998v1 Announce Type: cross Abstract: Large Language Model (LLM) alignment trains an LLM using preference data to produce outputs that better meet established quality standards. While LLM

SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing

arXiv:2606.29887v1 Announce Type: new Abstract: In real-world applications, guardrails are often expected to identify unsafe user-model interactions according to application-specific safety policies,

Semantic-Driven Scale and Spatial Selection for Efficient Cross-Modal Alignment in Referring Remote Sensing Image Segmentation

arXiv:2606.30244v1 Announce Type: new Abstract: Referring Remote Sensing Image Segmentation (RRSIS) seeks to localize and segment the target object or region specified by a natural language expression

Symbolic Mechanistic Data Attribution: Tracing Training Influence to Learned Behavioral Policies

arXiv:2606.29171v1 Announce Type: cross Abstract: While existing data attribution methods can identify which training examples build specific mechanistic circuits, they cannot explain how training dat

TF-MoE: Time-Frequency Mixture-of-Experts for Efficient Speech Separation

arXiv:2606.29575v1 Announce Type: cross Abstract: Recent advances in speech separation (SS) have led to compact front-end models with small parameter sizes, yet their high computational cost remains a

Research

The Fundamental Limits of Valid Transport Map Estimation

arXiv:2606.30574v1 Announce Type: new Abstract: Many modern generative modeling methods, including diffusion models, normalizing flows, and flow matching, estimate transport maps or plans between dist

researcharxiv-cs-lg

Towards Physical Intuitions for Alignment Dynamics: A Case Study With Randomness Crystallization

arXiv:2606.29933v1 Announce Type: new Abstract: The alignment of language models is typically studied through the lens of capability benchmarks, but the dynamics of how models change during post-train

safetyarxiv-cs-cl

TriageRA-CCF: Source-Side Clinical Confidence and Coverage Signals for Adaptive Rank Budgeting in Medical LLMs

arXiv:2606.29375v1 Announce Type: new Abstract: Medical large language models are commonly adapted with a fixed low-rank budget, even though medical questions differ substantially in confidence, clini

UniGP: Taming Diffusion Transformer for Prior-Preserved Unified Generation and Perception

arXiv:2606.30332v1 Announce Type: new Abstract: Recent advances in diffusion models have shown impressive performance in controllable image generation and dense prediction tasks. However, existing app

safetyarxiv-cs-cv

Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare

arXiv:2606.28666v1 Announce Type: cross Abstract: Agent-based AI has enabled the automation of tasks by exposing application tools and resources to large language models (LLMs). However, to improve sc

You Only Touch Once: 6-DoF Object Pose Estimation from Single Tactile Contact

model-releasesarxiv-cs-ro

arXiv:2606.28899v1 Announce Type: new Abstract: Accurate 6-DoF object pose estimation is fundamental to robotic manipulation, yet vision-based methods often fail under occlusion, poor lighting, and re

A Comparison of Fusion Techniques for Multi-Modal Human Activity Recognition on the HARMES Dataset

arXiv:2606.27886v1 Announce Type: new Abstract: Recent advances in Human Activity Recognition (HAR) from wearable sensors have shown that multi-modal deep learning models consistently outperform their

Accelerating Attention with Basis Decomposition

arXiv:2510.01718v2 Announce Type: replace Abstract: Attention is a core operation in large language models (LLMs). We present BD Attention (BDA), a lossless algorithmic reformulation of attention. BDA

Alibaba allegedly ran 28.8 million fraudulent API exchanges across 25,000 fake accounts to steal Claude's intelligence. If confirmed, it's t…

model-releasesemad-mostaque--x

Alibaba allegedly ran 28.8 million fraudulent API exchanges across 25,000 fake accounts to steal Claude's intelligence. If confirmed, it's the largest AI model theft ever attempted. The same week, the

Applicability of memorization indicators for early spotting of overfitting while recalibrating sEMG-decoders on low sample sizes

arXiv:2606.27855v1 Announce Type: cross Abstract: Deep learning models for surface electromyography (sEMG) can benefit substantially from subject-specific (re-)calibration, since no sufficiently large

Claude in Microsoft Foundry is now generally available, hosted on Azure. Azure customers get Claude Opus 4.8 and Claude Haiku 4.5, with Azur…

model-releasesboris-cherny--x

Claude models are now generally available through Microsoft Foundry on Azure infrastructure, providing Azure customers access to Claude Opus 4.8 and Claude Haiku 4.5. This integration allows enterpris

Cloud CISO Perspectives: How Google Cloud Security uses AI internally

model-releasesgoogle-cloud-ai

Welcome to the second Cloud CISO Perspectives for June 2026. Today, we’re discussing how we use AI to chart a path to autonomous software development lifecycle security.As with all Cloud CISO Perspect

Contagion Networks: Evaluator Preference Propagation in Multi-Agent LLM Systems

arXiv:2606.20493v2 Announce Type: replace-cross Abstract: When large language models serve as evaluators in multi-agent systems, their strategy preferences -- whether induced by explicit prompts or by

Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecasting

arXiv:2606.27688v1 Announce Type: cross Abstract: In financial forecasting, predictive performance depends not only on which model is trained, but also on how the trained model is deployed. We study t

safetyarxiv-cs-ai

Estimation--Prediction Tradeoff in Causal Probabilistic Temporal Graphs

arXiv:2606.28225v1 Announce Type: new Abstract: Temporal link prediction is usually evaluated by predictive performance on unseen edges, but in probabilistic temporal graphs this criterion can conflat

EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning

arXiv:2603.09731v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) are increasingly considered as a foundation for embodied agents, yet it remains unclear whether they

Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMs

arXiv:2606.27378v1 Announce Type: new Abstract: We introduce an axiomatic evaluation framework for latent thought representations in LLMs, comprising metrics that are independent of downstream benchma

From Detection to Action: Using LLM Agents for Fault-Tolerant Control

arXiv:2606.28011v1 Announce Type: cross Abstract: We propose an agentic Large Language Model (LLM) framework for active Fault-Tolerant Control (FTC) that transforms fault detection outputs into constr

Agents

GraphPilot: Grounded Scene Graph Conditioning for Language-Based Autonomous Driving

arXiv:2511.11266v4 Announce Type: replace Abstract: Vision-language models have recently emerged as promising planners for autonomous driving, where success hinges on topology-aware reasoning over spa

agentsarxiv-cs-cv

Learning to Evict from Key-Value Cache

arXiv:2602.10238v2 Announce Type: replace Abstract: The growing size of Large Language Models (LLMs) makes efficient inference challenging, primarily due to the memory demands of the autoregressive Ke

OrthoTryOn: Geometric Orthogonalization for Conflict-Free Unified Fashion Generation

arXiv:2606.27880v1 Announce Type: new Abstract: Unified fashion generation integrates tasks like virtual try-on and garment reconstruction into a single model to reduce task-specific adaptation costs.

Output-Space Allocation Costs for Calibration-Guided LLM Compression: An Empirical Study

arXiv:2606.27785v1 Announce Type: cross Abstract: Training-free compression methods for large language models (LLMs) often use calibration data to guide compression decisions. ROCKET, a recent method

Position Bias Correction is Insufficient for One-Pass Attention Sorting

arXiv:2606.27793v1 Announce Type: cross Abstract: Long-context language models suffer from position bias, where information in middle positions is underutilized. Attention Sorting addresses this by it

Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting

arXiv:2510.18874v3 Announce Type: replace-cross Abstract: Adapting language models (LMs) to new tasks via post-training carries the risk of degrading existing capabilities -- a phenomenon classically

Triadic Werewolf: A Jester Role for Multi-Hop Theory of Mind in LLMs

arXiv:2606.27909v1 Announce Type: cross Abstract: Theory-of-mind evaluations of large language models typically use dyadic social-deduction games, where every observable cue points to a single hidden

When Search Agents Should Ask: DiscoBench for Clarification-Aware Deep Search

arXiv:2606.27669v1 Announce Type: new Abstract: Search agents powered by large language models (LLMs) are increasingly used to solve complex information-seeking tasks, requiring multi-step retrieval a

ZooClaw-FashionSigLIP2: Distilled Fine-tuning for Robust Fashion Retrieval

arXiv:2606.27708v1 Announce Type: new Abstract: Adapting a foundation vision-language encoder to a specialized retrieval task creates a fundamental tradeoff: gains on the target distribution come at t

China’s Z.ai claims it can match Mythos on cybersecurity

model-releasesthe-verge-ai

China's Zhipu AI (Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from

28 Jun 2026

Elon Musk turns 55 today. Here are 55 milestones. Age 54: world's first trillionaire Age 54: takes SpaceX public Age 54: SpaceX acquires xAI…

model-releaseselon-musk--x

Elon Musk turns 55 today. Here are 55 milestones. Age 54: world's first trillionaire Age 54: takes SpaceX public Age 54: SpaceX acquires xAI Age 54: releases Grok 4 Age 53: launches Robotaxi service A

28 Jun 2026

https://huggingface.co/collections/deepseek-ai/deepspec

model-releasesclem-delangue--x

https://huggingface.co/collections/deepseek-ai/deepspec Good guy DeepSeek gives us accelerated models The most interesting one here is Gemma4-12B, I presume vision included. Might be the best local mo

28 Jun 2026

[AINews] OpenAI GPT-5.6 Sol / Terra / Luna — restricted to trusted partners

model-releaseslatent-space

OpenAI has released GPT-5.6 with three variants (Sol, Terra, Luna) in a restricted beta program limited to trusted partners, according to reporting from Latent Space. The restricted access model sugge

27 Jun 2026

A probabilistic framework for online test-time adaptation

arXiv:2606.26457v1 Announce Type: cross Abstract: This paper presents a probabilistic framework for online test-time adaptation problems. In them, a model is trained on labeled data but must adapt to

26 Jun 2026

Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents

arXiv:2606.26479v1 Announce Type: cross Abstract: Recent work (2024 to 2026) has converged on a strategy for defending tool-using LLM agents against indirect prompt injection: rather than training the

26 Jun 2026

Comparing BERT Sentence-Pair Classification and Few-Shot LLM Prompting for Detecting Threat and Solution Framing in German Climate News

arXiv:2606.26489v1 Announce Type: new Abstract: News media play a central role in shaping public perceptions of climate change, and whether coverage emphasizes threats or solutions has measurable effe