All entries

Asymptotically Log-Optimal Bayes-Assisted Confidence Sequences for Bounded Means

arXiv:2605.07964v1 Announce Type: cross Abstract: Confidence sequences based on test martingales provide time-uniform uncertainty quantification for the mean of bounded IID observations without parame

researcharxiv-cs-lg

AsyncEvGS: Asynchronous Event-Assisted Gaussian Splatting for Handheld Motion-Blurred Scenes

DGX agent

arXiv:2605.07192v1 Announce Type: new Abstract: 3D reconstruction methods such as 3D Gaussian Splatting (3DGS) and Neural Radiance Fields (NeRF) achieve impressive photorealism but fail when input ima

researcharxiv-cs-cv

11 May 2026

AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models

arXiv:2605.07308v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have significantly advanced the capabilities of robotic agents in executing diverse tasks; however, they still face

applicationsarxiv-cs-ro

Agents

ATHENA: Agentic Team for Hierarchical Evolutionary Numerical Algorithms

arXiv:2512.03476v2 Announce Type: replace-cross Abstract: Bridging the gap between theoretical conceptualization and computational implementation is a major bottleneck in Scientific Computing (SciC) a

agentsarxiv-cs-ai

Attention Sparsity is Input-Stable: Training-Free Sparse Attention for Video Generation via Offline Sparsity Profiling and Online QK Co-Clustering

arXiv:2603.18636v2 Announce Type: replace Abstract: Diffusion Transformers (DiTs) achieve strong video generation quality but suffer from high inference cost due to dense 3D attention, motivating spar

researcharxiv-cs-cv

Attention Transfer Is Not Universally Effective for Vision Transformers

model-releasesarxiv-cs-cv

arXiv:2605.07191v1 Announce Type: new Abstract: A recent work shows that Attention Transfer, which transfers only the attention patterns from a pre-trained teacher Vision Transformer (ViT) to a random

Attribution-Based Neuron Utility for Plasticity Restoration in Deep Networks

model-releasesarxiv-cs-lg

arXiv:2605.06834v1 Announce Type: new Abstract: Continual learning research attempts to conserve two fundamental capabilities: new knowledge acquisition and the preservation of previously acquired kno

Industry

Audi has a new Q9 flagship coming soon: Here's its interior

Audi's new Q9 flagship features a spacious, exclusive interior conceived as a sanctuary, offering up to seven seats with electrically powered doors and a large panoramic sunroof with switchable transp

industryars-technica

AudioFace: Language-Assisted Speech-Driven Facial Animation with Multimodal Language Models

arXiv:2605.07478v1 Announce Type: new Abstract: Speech-driven facial animation requires accurate correspondence between acoustic signals and facial motion, especially for articulation-related mouth mo

applicationsarxiv-cs-cv

Tools

Automate progressive rollouts with Vercel Flags

Vercel Flags enables automated progressive rollouts, allowing developers to gradually deploy features to increasing percentages of users rather than releasing to everyone simultaneously. This approach

toolsvercel-blog

Automate security detection, validation, and response with Daybreak

Daybreak is an automated security system that detects potential security threats, validates their legitimacy, and executes response actions with minimal human intervention. Developed by OpenAI, it rep

model-releasesopenai--x

Automated Evaluation can Distinguish the Good and Bad AI Responses to Patient Questions about Hospitalization

arXiv:2510.00436v2 Announce Type: replace Abstract: Automated approaches to answer patient-posed health questions are rising, but selecting among systems requires reliable evaluation. The current gold

researcharxiv-cs-ai

Automatic Image-Level Morphological Trait Annotation for Organismal Images

arXiv:2604.01619v3 Announce Type: replace-cross Abstract: Morphological traits are physical characteristics of biological organisms that provide vital clues on how organisms interact with their enviro

researcharxiv-cs-ai

b9102: [SYCL] Add OP im2col_3d (#22903)

Release b9102 adds support for the im2col_3d operator to the SYCL backend in llama.cpp, expanding GPU-accelerated tensor operations for Intel GPU inference. The im2col_3d operation converts 3D input v

b9103

B9103 is a recent release of llama.cpp , a C/C++ implementation for large language model inference. This release represents one of the latest versions in the project's active development cycle, buildi

b9105

Release b9105 of llama.cpp includes updates to the AllReduce implementation for CUDA, introducing a NCCL-free provider for tensor parallelism that pipelines data transfers and GPU reduction operations

b9106

Release b9106 of llama.cpp includes support for asymmetric FlashAttention in scalar, MMQ, and cooperative matrix paths for Vulkan. The release provides builds for multiple platforms including macOS, L

b9109

Release b9109 of llama.cpp introduces refinements to the CUDA tensor parallelism AllReduce implementation, including renaming the --allreduce flag to --reduction-provider and updates to NCCL-free AllR

b9110

The search results show recent llama.cpp releases and general information but don't contain specific details about the b9110 release. Based on the search patterns and similar recent releases documente

b9112

Release b9112 of llama.cpp introduces a NCCL-free AllReduce implementation for LLAMA_SPLIT_MODE_TENSOR using a single-phase CUDA kernel, and adds an --allreduce flag to llama-bench to select between A

BalCapRL: A Balanced Framework for RL-Based MLLM Image Captioning

arXiv:2605.07394v1 Announce Type: cross Abstract: Image captioning is one of the most fundamental tasks in computer vision. Owing to its open-ended nature, it has received significant attention in the

researcharxiv-cs-ai

Bare Metal: Z-Image Turbo - Flux.2 Klein 9b - Wan 2.2

local-air-stablediffusion

Bare Metal: Z-Image Turbo - Flux.2 Klein 9b - Wan 2.2 discusses comparisons between these lightweight AI image generation models, with Flux.2 Klein 9B showing good prompt adherence and natural composi

Bayesian Attention Mechanism: A Probabilistic Framework for Positional Encoding and Context Length Extrapolation

arXiv:2505.22842v4 Announce Type: replace Abstract: Transformer-based language models rely on positional encoding (PE) to handle token order and support context length extrapolation. However, existing

researcharxiv-cs-cl

Bayesian Conformal Prediction as a Decision Risk Problem

arXiv:2602.03331v2 Announce Type: replace Abstract: We propose Bayesian Conformal Prediction (BCP), a framework that combines Bayesian posterior predictive distributions with PAC-style conformal risk

researcharxiv-cs-lg

Bayesian Fine-tuning in Projected Subspaces

model-releasesarxiv-cs-lg

arXiv:2605.07706v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning of large models by decomposing weight updates into low-rank matrices, significantly r

Bayesian Sensitivity of Causal Inference Estimators under Evidence-Based Priors

arXiv:2605.07993v1 Announce Type: new Abstract: Causal inference, especially in observational studies, relies on untestable assumptions about the true data-generating process. Sensitivity analysis hel

applicationsarxiv-cs-lg

BEAVER: An Efficient Deterministic LLM Verifier

arXiv:2512.05439v2 Announce Type: replace Abstract: As large language models (LLMs) transition from research prototypes to production systems, practitioners often need reliable methods to verify model

safetyarxiv-cs-ai

BeeVe: Unsupervised Acoustic State Discovery in Honey Bee Buzzing

arXiv:2605.07903v1 Announce Type: cross Abstract: Discovering structure in biological signals without supervision is a fundamental problem in computational intelligence, yet existing bioacoustic metho

applicationsarxiv-cs-ai

Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight

arXiv:2605.07021v1 Announce Type: new Abstract: Reasoning in Large Language Models (LLMs) poses a challenge for oversight as many misaligned behaviors do not surface until reasoning concludes. To addr

Agents

Belief Memory: Agent Memory Under Partial Observability

arXiv:2605.05583v2 Announce Type: replace Abstract: LLM agents that operate over long context depend on external memory to accumulate knowledge over time. However, existing methods typically store eac

agentsarxiv-cs-ai

Bellman Calibration for V-Learning in Offline Reinforcement Learning

arXiv:2512.23694v2 Announce Type: replace-cross Abstract: Reliable long-horizon value prediction is difficult in offline reinforcement learning because fitted value methods combine bootstrapping, func

safetyarxiv-cs-lg

Benchmarked Yet Not Measured -- Generative AI Should be Evaluated Against Real-World Utility

model-releasesarxiv-cs-cl

arXiv:2605.06856v1 Announce Type: cross Abstract: Generative AI systems achieve impressive performance on standard benchmarks yet fail to deliver real-world utility, a disconnect we identify across 28

Benchmarking EngGPT2-16B-A3B against Comparable Italian and International Open-source LLMs

arXiv:2605.07731v1 Announce Type: cross Abstract: This report benchmarks the performance of ENGINEERING Ingegneria Informatica S.p.A.'s EngGPT2MoE-16B-A3B LLM, a 16B parameter Mixture of Experts (MoE)

Benchmarking Foundation Models for Renal Lesion Stratification in CT

model-releasesarxiv-cs-cv

arXiv:2605.07749v1 Announce Type: new Abstract: The rapid proliferation of open-source medical foundation models (FMs) raises a practical question: how well do their pre-trained representations transf

Benchmarking World-Model Learning with Environment-Level Queries

arXiv:2510.19788v4 Announce Type: replace Abstract: World models are central to building AI agents capable of flexible reasoning and planning. Yet current evaluations (i) test only properties measurab

Best LLM on a 32Gb M5 MBA

With 32GB of RAM, 30B Q4 models run comfortably on M5 MacBook Air , and popular choices include Qwen 3.6 27B (requiring 18GB+ VRAM) and Qwen 3.6 9B (lightweight, runs on 8GB) . Ollama's macOS implemen

local-air-ollama

Better Protein Function Prediction by Modeling Survivorship Bias

arXiv:2605.06879v1 Announce Type: new Abstract: Protein sequence data from nature exhibits survivorship bias: we only observe data from those organisms that survive and reproduce, while non-functional

safetyarxiv-cs-lg

Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs

arXiv:2605.07806v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used in settings where reliable self-assessment is critical. Assessing model reliability has evolved fro

safetyarxiv-cs-ai

Beyond Defenses: Manifold-Aligned Regularization for Intrinsic 3D Point Cloud Robustness

arXiv:2605.07590v1 Announce Type: new Abstract: Despite extensive progress in point cloud robustness, existing methods primarily improve performance through augmentation or defense mechanisms, while o

researcharxiv-cs-cv

Beyond Distribution Estimation: Simplex Anchored Structural Inference Towards Universal Semi-Supervised Learning

arXiv:2605.07557v1 Announce Type: new Abstract: Semi-supervised learning faces significant challenges in realistic scenarios where labeled data is scarce and unlabeled data follows unknown, arbitrary

researcharxiv-cs-lg

Beyond Factor Aggregation: Gauge-Aware Low-Rank Server Representations for Federated LoRA

arXiv:2605.06733v1 Announce Type: cross Abstract: Federated LoRA enables parameter-efficient adaptation of large language models under decentralized data and limited client resources.However, directly

Beyond Factual Accuracy: Evaluating Global Reasoning Integrity in RAG Systems with LogicScore

model-releasesarxiv-cs-cl

arXiv:2601.15050v4 Announce Type: replace Abstract: Current evaluation methods for Retrieval Augmented Generation (RAG) suffer from extit{factual myopia}: they relentlessly emphasize factual accuracy

Beyond GSD-as-Token: Continuous Scale Conditioning for Remote Sensing VLMs

model-releasesarxiv-cs-cv

arXiv:2605.07562v1 Announce Type: new Abstract: Remote sensing vision-language models (RS-VLMs) face a fundamental mismatch with natural-image counterparts: the same geographic object exhibits radical

Beyond 'I cannot fulfill this request': Alleviating Rigid Rejection in LLMs via Label Enhancement

arXiv:2605.07883v1 Announce Type: new Abstract: Large Language Models (LLMs) rely on safety alignment to obey safe requests while refusing harmful ones. However, traditional refusal mechanisms often l

safetyarxiv-cs-cl

Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning

model-releasesarxiv-cs-lg

arXiv:2605.07333v1 Announce Type: new Abstract: In-context reinforcement learning (ICRL) studies agents that, after pretraining, adapt to new tasks by conditioning on additional context without parame

Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation

arXiv:2605.07111v1 Announce Type: cross Abstract: Recent literature on fine-tuning Large Language Models highlights a fundamental debate. While Full Fine-Tuning (FFT) provides the representational pla

Beyond Pairs: Your Language Model is Secretly Optimizing a Preference Graph

arXiv:2605.08037v1 Announce Type: cross Abstract: Direct Preference Optimization (DPO) aligns language models using pairwise preference comparisons, offering a simple and effective alternative to Rein

safetyarxiv-cs-ai

Beyond Reasoning: Reinforcement Learning Unlocks Parametric Knowledge in LLMs

arXiv:2605.07153v1 Announce Type: new Abstract: Reinforcement learning (RL) has achieved remarkable success in LLM reasoning, but whether it can also improve direct recall of parametric knowledge rema

researcharxiv-cs-cl