Search: “reasoning”

Counterfactual Tests for Measuring Chain-of-Thought Faithfulness in Visual Language Models

arXiv:2609.06704v1 Announce Type: new Abstract: Chain-of-thought (CoT) may often look plausible, yet it may not faithfully reflect the model's decision-making process. While methods for measuring the

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

CrACK: Adversarial Attacks on Cross-Model Consistency in Collaborative Vision Foundation Models

DGX agent

arXiv:2609.07499v1 Announce Type: new Abstract: Training-free collaborative pipelines that integrate Vision Foundation Models such as CLIP, SAM, and DINO achieve strong open-vocabulary dense predictio

safetyarxiv-cs-cv

9 Sept 2026

CrowdTraj: A Benchmark for Dense Crowd Trajectory Prediction in Realistic Crowded Environments

arXiv:2609.07685v1 Announce Type: new Abstract: In real-world applications, pedestrian trajectory prediction models rely on inputs from detection and tracking systems. Prior trajectory prediction benc

Deploying Qwen3.8-2.4T-A95B on Amazon SageMaker HyperPod with vLLM

model-releasesaws-ml-blog

Learn how to deploy Qwen3.8-2.4T-A95B, a 2.4-trillion-parameter open-weight model, on Amazon SageMaker HyperPod with vLLM. This walkthrough covers cluster provisioning, NVFP4 quantization, and an Open

Distillation as Probability Transport: Routed On-Policy Distillation

arXiv:2609.08337v1 Announce Type: new Abstract: On-policy distillation (OPD) transfers teacher knowledge on student-generated trajectories, but efficient sampled objectives reduce the teacher distribu

DIVA: Exploiting Cross-Step Conditional Propagation for Visual Jailbreaks in Discrete Diffusion Vision-Language Models

arXiv:2609.05525v1 Announce Type: new Abstract: Large vision-language models (VLMs) are increasingly deployed in safety-critical settings, yet existing visual jailbreak research has focused almost exc

safetyarxiv-cs-cv

Local Ai

Do Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?

arXiv:2609.08331v1 Announce Type: new Abstract: Video Large Language Models (VideoLLMs) are increasingly deployed in safety-critical applications such as content moderation and video analytics. To pro

local-aiarxiv-cs-cv

Do Quantum AIs Dream in Paths? Path-Integral Slow Thinking through Grover Interference

model-releasesarxiv-cs-lg

arXiv:2609.05842v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards enables large language models to think slowly, but the same training can induce policy collapse: probab

Local Ai

Does ollama phone home?

im new to using ai so i may be missing something. I set up ollama but dont want any of my data to go to any of the model providers, for certain models i appear to need an api key to download the model

local-air-ollama

Don't let FOMO win if you're interested in local llm from a hobby/learning aspect

model-releasesr-localllama

Just a reminder for those out there itching to get into local llms - don't let FOMO or 'gear acquisition syndrom' take over. No matter the hobby, it's so easy to get stuck in a trap where we buy more

Dreaming in Flow: Generative Grounding Feedback for Self-Evolving Unified Multimodal Models

arXiv:2609.08282v1 Announce Type: new Abstract: Unified multimodal models integrate visual understanding and generation within a single network, yet the two capabilities are commonly optimized as sepa

Dual-Layer Semantic-Spatial Belief Mapping for Aerial Object Goal Navigation

model-releasesarxiv-cs-ro

arXiv:2609.08164v1 Announce Type: new Abstract: Aerial Object Goal Navigation (ObjectNav) requires an unmanned aerial vehicle (UAV) to locate a described target in an unknown outdoor environment using

Evaluation Principles for MRI-MRA Registration in Trigeminal Neuralgia: An ROI-Centered Neurovascular Benchmark

arXiv:2609.08805v1 Announce Type: new Abstract: Preoperative evaluation of trigeminal neuralgia (TN) often requires joint interpretation of structural MRI, which depicts the trigeminal nerve and surro

FedSubMuon: Communication-Efficient Federated LLM Fine-Tuning via Structured Subspace Muon

model-releasesarxiv-cs-lg

arXiv:2609.06073v1 Announce Type: new Abstract: Federated fine-tuning adapts large language models (LLMs) to decentralized client data, but its scalability in cross-device training is often limited by

GLM 5.3 Flash Q4 @ 60tps / 550tps on M3 Ultra

model-releasesr-localllama

I have been a dwarfstar fan for awhile and I really liked glm 5.3 flash but needed it to be materially faster to feel good using it. In the screenshot you can see the outcome of using the model with a

GoDeep: Annotation-Free Open-Vocabulary 3D Scene Understanding via Language-Space Lifting

arXiv:2609.09082v1 Announce Type: new Abstract: Open vocabulary 3D semantic segmentation methods typically lift CLIP features into 3D. This embeds points in a joint vision-language space known to beha

HB-PVI: A Hierarchical Bayesian Personalization and Value-of-Information Framework for Complex Activity Recognition

arXiv:2609.05582v1 Announce Type: new Abstract: Personalization can improve activity-recognition performance, but participant-specific gains are heterogeneous, and every additional calibration label h

How Well Can Modern LLMs Act as Agent Cores in Radiology Environments?

arXiv:2412.09529v4 Announce Type: replace Abstract: Radiology, with its heterogeneous modalities, anatomies, and evolving protocols, is a natural yet high-stakes testbed for agentic AI. As LLMs grow m

ID-Align: RoPE-Conscious Position Remapping for Dynamic High-Resolution Adaptation in Vision-Language Models

arXiv:2505.21465v2 Announce Type: replace Abstract: Currently, a prevalent approach for enhancing Vision-Language Models (VLMs) performance is to encode both the high-resolution version and the thumbn

If they truly believe the risk is that high, and I think they do, then we need 100x more research and understanding of it. That requires the…

If they truly believe the risk is that high, and I think they do, then we need 100x more research and understanding of it. That requires them to openly share their models, datasets, training code, and

safetyclem-delangue--x

Kairos: A Dataset for Fine-Grained Video-Language Modeling over Space, Time, and Dynamics

arXiv:2609.08755v1 Announce Type: new Abstract: Many emerging video language modeling tasks require systems to move beyond clip-level abstraction and model visual content as it unfolds over extended t

safetyarxiv-cs-cv

Learning to Use Imagination: Progress-Conditioned Future Utilization for World Action Models

arXiv:2609.06578v1 Announce Type: new Abstract: World Action Models (WAMs) extend Vision-Language-Action (VLA) models by incorporating future visual dynamics into action generation. However, existing

Let It Go or Learn to Self-Correct: Continuous Diffusion for Constrained Discrete Tasks

arXiv:2609.09009v1 Announce Type: new Abstract: Denoising Diffusion Probabilistic Models (DDPMs) generate samples by starting from noise and repeatedly denoising while keeping each update close to the

Leveraging Visual and Geometric Priors for Metric-scale and Complete Vehicle Gaussian Reconstruction from Limited Views

arXiv:2609.08841v1 Announce Type: new Abstract: High-fidelity vehicle assets are essential for controllable traffic scene generation, particularly for synthesizing rare and safety-critical long-tail s

safetyarxiv-cs-cv

Linear Algebra Foundations of Efficient Attention: A Phase Reversal in Rank Collapse Under SVD Compression

arXiv:2609.06341v1 Announce Type: new Abstract: Linear algebra provides the framework of concepts (matrix rank, singular value decomposition (SVD), and eigendecomposition) that modern artificial intel

Long-Horizon Language Model Reinforcement Learning via Progressive Point Matching

arXiv:2609.07303v1 Announce Type: new Abstract: Current paradigms for training language models via reinforcement learning rely heavily on sparse outcome rewards. However, as we pursue tasks that requi

LWiAI Podcast #256 - Fable 5.1, Astra Tease, Gemini 3.8 Flash

model-releaseslast-week-in-ai

The 256th episode of the LWiAI Podcast (recorded 09/03/2026) discusses major AI developments from the prior week: Anthropic’s Claude Fable 5.1 and Mythos 5.1, which feature lower pricing, stronger age

MetaKV: Adaptive KV Cache Compression for Constrained LLM Inference

arXiv:2609.07966v1 Announce Type: new Abstract: Key--value (KV) cache compression is an effective way to reduce the memory overhead of large language model (LLM) inference, particularly for long-conte

Agents

MiMo-X-Pro-Preview and MiMo-X-Flash-Preview - New Mimo Model found in Mimo Desktop preview announcement

From the email 'Early access to new models During the program, approved users can try two next-generation Xiaomi MiMo models for a limited time: MiMo-X-Pro-Preview and MiMo-X-Flash-Preview. The MiMo-X

agentsr-localllama

Tutorials

Model Predictive Control of Tensegrity Robots via Contact-Aware Graph Neural Dynamics Model

arXiv:2609.08958v1 Announce Type: new Abstract: Tensegrity robots offer lightweight, compliant mobility over challenging terrain but remain difficult to model and control due to complex contact-rich d

tutorialsarxiv-cs-ro

MoEMB: Scaling Universal Multimodal Embeddings with Efficient Mixture-of-Experts Models

arXiv:2609.08663v1 Announce Type: new Abstract: Universal multimodal embedding (UME) increasingly demands encoder's capacity for handling a broad range of tasks and modalities with increased complexit

Multi-Grid Post-Training for Long-Form Multi-Shot Video Generation

arXiv:2609.06373v1 Announce Type: new Abstract: Generating long-form multi-shot videos requires coherent within-shot motion and visually consistent narratives across shots. Existing video generators f

Observe Before You Alert: Adaptive Driver Alerting with Vision-Language Models

arXiv:2609.08130v1 Announce Type: new Abstract: Driver alerting from dashcam video requires sequential decision-making under partial observability: a system must decide not only whether a scene is ris

Agents

Omni Interaction Agent Technical Report

arXiv:2609.08977v1 Announce Type: cross Abstract: In this work, we present Gander, an end-to-end model that unifies omni perception, realtime interaction, and agentic capabilities within a single fram

agentsarxiv-cs-lg

One Rate Is Not Enough: Adaptive Anisotropic Learning Rates for LoRA Fine-Tuning

model-releasesarxiv-cs-lg

arXiv:2609.05885v1 Announce Type: new Abstract: Low-rank adaptation (LoRA) has become the standard for parameter-efficient fine-tuning of large language models. Most LoRA variants follow a uniform-LR

Physico-Geospatial Grounded Scene Interpretation for Mobile Robotics

arXiv:2609.06629v1 Announce Type: new Abstract: Recent advancements in deep learning allow robotic agents to interact with dynamic and unstructured environments. Of special interest is the integration

ReMoMask-2: Latent Retrieval-Augmented Masked Motion Generation

arXiv:2609.08365v1 Announce Type: new Abstract: Text-to-motion (T2M) generation maps natural language to human joint movements, aiding gaming, VR, and robotics. Retrieval-Augmented Text-to-Motion (RAG

Running qwen 3.8 27B iq3 xxs on RTX 3060.

model-releasesr-localllama

Getting anywhere from 10 - 20 tps. Thinking Off . Took about 4 mins and 7 mins. Running on about 'IQ3_S - 3.4375 bpw' 37.03.960.932 I slot print_timing: id 0 | task 2665 | prompt processing, n_tokens

STP-BENCH: A Unified Systematic Benchmark for Virtual Spatial Transcriptomics from Histopathology Images

arXiv:2609.05956v1 Announce Type: new Abstract: Spatial transcriptomics (ST) provides unprecedented insights into tumor heterogeneity by capturing spatially resolved gene expression, yet its high expe

TASG-Explore: Traversability-Aware Sector-Guided Exploration for Ground Robot on Uneven Terrain

model-releasesarxiv-cs-ro

arXiv:2609.08512v1 Announce Type: new Abstract: Autonomous exploration on uneven terrain requires ground robots to balance exploration efficiency, coverage completeness, and terrain safety. Detailed t

The case for boycotting Generative AI: https://garymarcus.substack.com/p/the-case-for-boycotting-generative

Gary Marcus argues that current generative‑AI models fail to meet basic standards of reliability because they routinely hallucinate facts, amplify existing biases, and provide no transparent audit tra

safetygary-marcus--x

Tracking the Moving Frontier: Long-Short Term Advantage Estimator

arXiv:2609.06671v1 Announce Type: new Abstract: Group-based RLVR methods estimate advantages by repeatedly sampling multiple trajectories for each prompt, making long-horizon agent training expensive

Tri-PvP: Exposing Modality Bias in Omni-Modal Large Language Models through Perceptual-Propositional Evidence Conflicts

arXiv:2609.06011v1 Announce Type: cross Abstract: Omni-modal large language models (OLLMs) jointly process vision, audio, and text, yet their modality bias under cross-modal conflict remains underexpl

Applications

TriCCOT: Tri-part Convolutional Conformal Transformer for Onboard Space Object Detection

arXiv:2609.08659v1 Announce Type: new Abstract: Onboard object detection in Earth observation is constrained by limited computational resources and the absence of fully corrected imagery. While convol

applicationsarxiv-cs-cv

TRIUNE-Net: Harmonizing Scale, Shape, and Efficiency in Pancreatic Tumor Segmentation

arXiv:2609.08303v1 Announce Type: new Abstract: Pancreatic tumor segmentation in 3D CT volumes is challenged by extreme scale variability across both the pancreas and tumor, and highly irregular tumor

Agents

TrojanWorld: Backdooring World-Model Agents via Imagination Steering

arXiv:2609.07051v1 Announce Type: new Abstract: World models increasingly serve as the predictive core of model-based reinforcement learning agents, enabling them to simulate future dynamics and reaso

agentsarxiv-cs-lg

Unexplored flaws in multiple-choice VQA make benchmarking unreliable

arXiv:2511.22341v2 Announce Type: replace Abstract: Previous works identify sensitivity to option order as a key issue in multiple-choice VQA (MC-VQA) evaluation and propose protocols to mitigate this

Uni-Inter: Unifying 3D Human Motion Synthesis Across Diverse Interaction Contexts

arXiv:2511.13032v2 Announce Type: replace Abstract: We present Uni-Inter, a unified framework for human motion generation that supports a wide range of interaction scenarios: including human-human, hu