Search: “models”

Demis says he wants to see a Western open source AI stack and that we’re losing to China. He also says Google doesn’t have enough compute to…

model-releasesclem-delangue--x

Demis says he wants to see a Western open source AI stack and that we’re losing to China. He also says Google doesn’t have enough compute to build two frontier (open and closed) models, which is why G

Safety

EvoSelect: Data-Efficient LLM Evolution for Targeted Task Adaptation

DGX agent

arXiv:2604.26170v1 Announce Type: new Abstract: Adapting large language models (LLMs) to a targeted task efficiently and effectively remains a fundamental challenge. Such adaptation often requires ite

safetyarxiv-cs-cl

30 Apr 2026

FaaSMoE: A Serverless Framework for Multi-Tenant Mixture-of-Experts Serving

arXiv:2604.26881v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) models offer high capacity with efficient inference cost by activating a small subset of expert models per input. However, de

researcharxiv-cs-lg

Human-in-the-Loop Benchmarking of Heterogeneous LLMs for Automated Competency Assessment in Secondary Level Mathematics

arXiv:2604.26607v1 Announce Type: new Abstract: As Competency-Based Education (CBE) is gaining traction around the world, the shift from marks-based assessment to qualitative competency mapping is a m

HumanOmni-Speaker: Identifying Who said What and When

arXiv:2603.21664v2 Announce Type: replace Abstract: While Omni-modal Large Language Models have made strides in joint sensory processing, they fundamentally struggle with a cornerstone of human intera

Learning Neural Operator Surrogates for the Black Hole Accretion Code

arXiv:2604.25985v1 Announce Type: cross Abstract: General-relativistic magnetohydrodynamic (GR-MHD) simulations are essential for studying black hole accretion, relativistic jets, and magnetic reconne

Pointer-CAD: Unifying B-Rep and Command Sequences via Pointer-based Edges & Faces Selection

arXiv:2603.04337v2 Announce Type: replace-cross Abstract: Constructing computer-aided design (CAD) models is labor-intensive but essential for engineering and manufacturing. Recent advances in Large L

applicationsarxiv-cs-cl

Local Ai

Privacy-Preserving Federated Learning Framework for Distributed Chemical Process Optimization

arXiv:2604.26073v1 Announce Type: cross Abstract: Industrial chemical plants often operate under strict data confidentiality constraints, making centralized data-driven process modeling difficult. Fed

local-aiarxiv-cs-ai

Reasoning Gets Harder for LLMs Inside A Dialogue

arXiv:2603.20133v2 Announce Type: replace Abstract: Large Language Models (LLMs) achieve strong performance on many reasoning benchmarks, yet these evaluations typically focus on isolated tasks that d

SciMDR: Advancing Scientific Multimodal Document Reasoning

arXiv:2603.12249v2 Announce Type: replace-cross Abstract: Constructing scientific multimodal document reasoning datasets for foundation model training involves an inherent trade-off among scale, faith

The Prompt Engineering Report Distilled: Quick Start Guide for Life Sciences

arXiv:2509.11295v2 Announce Type: replace Abstract: Developing effective prompts demands significant cognitive investment to generate reliable, high-quality responses from Large Language Models (LLMs)

Through a Compressed Lens: Investigating The Impact of Quantization on Factual Knowledge Recall

arXiv:2505.13963v3 Announce Type: replace Abstract: Quantization methods are widely used to accelerate inference and streamline the deployment of large language models (LLMs). Although quantization's

researcharxiv-cs-cl

tweeted about this yesterday and Cursor already dropped the alpha today! 🚀very cool to see how us, them, and others have converged on good …

model-releasesharrison-chase--x

tweeted about this yesterday and Cursor already dropped the alpha today! 🚀very cool to see how us, them, and others have converged on good design patterns in Agent + Harness Engineering: 1. Tuning dif

Value-Guided Iterative Refinement and the DIQ-H Benchmark for Evaluating VLM Robustness

arXiv:2512.03992v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) are essential for embodied AI and safety-critical applications, such as robotics and autonomous systems. However

Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses

arXiv:2604.25850v1 Announce Type: new Abstract: Harnesses have become a central determinant of coding-agent performance, shaping how models interact with repositories, tools, and execution environment

BARRED: Synthetic Training of Custom Policy Guardrails via Asymmetric Debate

arXiv:2604.25203v1 Announce Type: new Abstract: Deploying guardrails for custom policies remains challenging, as generic safety models fail to capture task-specific requirements, while prompting LLMs

safetyarxiv-cs-cl

BLASST: Dynamic BLocked Attention Sparsity via Softmax Thresholding

arXiv:2512.12087v3 Announce Type: replace Abstract: The growing demand for long-context inference capabilities in Large Language Models (LLMs) has intensified the computational and memory bottlenecks

Conditional Flow Matching for Probabilistic Downscaling of Maximum 3-day Snowfall in Alaska

arXiv:2604.25172v1 Announce Type: cross Abstract: Precipitation in complex terrain is governed by orographic processes operating at scales of a few kilometers, yet climate models typically run at reso

researcharxiv-cs-lg

Tutorials

DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing

arXiv:2604.25477v1 Announce Type: new Abstract: Recent image editing models have achieved strong visual fidelity but often struggle with tasks requiring complex reasoning. To investigate and enhance t

tutorialsarxiv-cs-cv

DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference

arXiv:2510.19669v4 Announce Type: replace Abstract: Recent reasoning Large Language Models (LLMs) demonstrate remarkable problem-solving abilities but often generate long thinking traces whose utility

researcharxiv-cs-cl

Doing More With Less: Revisiting the Effectiveness of LLM Pruning for Test-Time Scaling

arXiv:2604.25098v1 Announce Type: cross Abstract: While current Large Language Models (LLMs) exhibit remarkable reasoning capabilities through test-time compute scaling (TTS), their massive parameter

FED-FSTQ: Fisher-Guided Token Quantization for Communication-Efficient Federated Fine-Tuning of LLMs on Edge Devices

arXiv:2604.25421v1 Announce Type: new Abstract: Federated fine-tuning provides a practical route to adapt large language models (LLMs) on edge devices without centralizing private data, yet in mobile

Golden RPG: Confidence-Adaptive Region-Aware Noise for Compositional Text-to-Image Generation

arXiv:2604.25314v1 Announce Type: new Abstract: Compositional text-to-image (T2I) generation requires a model to honour multiple sub-prompts that describe distinct image regions. Recent work shows tha

I released LLM 0.32a0 this morning, a major backwards-compatible refactor of my LLM Python library and CLI tool for working with language mo…

model-releasessimon-willison--x

I released LLM 0.32a0 this morning, a major backwards-compatible refactor of my LLM Python library and CLI tool for working with language models - the new changes should help LLM work better with reas

Improving LLM Predictions via Inter-Layer Structural Encoders

arXiv:2603.22665v2 Announce Type: replace Abstract: The standard practice in Large Language Models (LLMs) is to base predictions on final-layer representations. However, intermediate layers encode com

Limited Linguistic Diversity in Embodied AI Datasets

arXiv:2601.03136v2 Announce Type: replace Abstract: Language plays a critical role in Vision-Language-Action (VLA) models, yet the linguistic characteristics of the datasets used to train and evaluate

researcharxiv-cs-cl

LLM-ReSum: A Framework for LLM Reflective Summarization through Self-Evaluation

arXiv:2604.25665v1 Announce Type: new Abstract: Reliable evaluation of large language model (LLM)-generated summaries remains an open challenge, particularly across heterogeneous domains and document

MICo-150K: A Comprehensive Dataset Advancing Multi-Image Composition

arXiv:2512.07348v2 Announce Type: replace Abstract: In controllable image generation, synthesizing coherent and consistent images from multiple reference inputs, i.e., Multi-Image Composition (MICo),

Quantifying and Mitigating Socially Desirable Responding in LLMs: A Desirability-Matched Graded Forced-Choice Psychometric Study

arXiv:2602.17262v2 Announce Type: replace Abstract: Human self-report questionnaires are increasingly used in NLP to benchmark and audit large language models (LLMs), from persona consistency to safet

RCProb: Probabilistic Rule Extraction for Efficient Simplification of Tree Ensembles

arXiv:2604.25304v1 Announce Type: new Abstract: Tree ensembles are widely used in industrial machine learning due to their strong predictive performance and efficient training procedures. However, as

Relational In-Context Learning via Synthetic Pre-training with Structural Prior

arXiv:2603.03805v2 Announce Type: replace Abstract: Relational Databases (RDBs) are the backbone of modern business, yet they lack foundation models comparable to those in text or vision. A key obstac

applicationsarxiv-cs-lg

ReSim: Reliable World Simulation for Autonomous Driving

arXiv:2506.09981v2 Announce Type: replace Abstract: How can we reliably simulate future driving scenarios under a wide range of ego driving behaviors? Recent driving world models, developed exclusivel

safetyarxiv-cs-cv

Sensitivity-Based Tube NMPC for Cooperative Aerial Structures Under Parametric Uncertainty

arXiv:2604.25766v1 Announce Type: new Abstract: This paper presents a sensitivity-based tube Nonlinear Model Predictive Control (NMPC) framework for cooperative aerial chains under bounded parametric

researcharxiv-cs-ro

TopoMamba: Topology-Aware Scanning and Fusion for Segmenting Heterogeneous Medical Visual Media

arXiv:2604.25545v1 Announce Type: new Abstract: Visual state-space models (SSMs) have shown strong potential for medical image segmentation, yet their effectiveness is often limited by two practical i

researcharxiv-cs-cv

When Thoughts Meet Facts: Reusable Reasoning for Long-Context LMs

arXiv:2510.07499v2 Announce Type: replace Abstract: Recent Long-Context Language Models (LCLMs) can process hundreds of thousands of tokens in a single prompt, enabling new opportunities for knowledge

Xiaomi MiMo-V2.5-Pro achieves multiple breakthroughs in the latest Arena rankings (Apr 26, 2026) 🔥 🏆 Text Arena (Expert) — #6 globally | #…

applicationsjeremy-howard--x

Xiaomi MiMo-V2.5-Pro achieves multiple breakthroughs in the latest Arena rankings (Apr 26, 2026) 🔥 🏆 Text Arena (Expert) — #6 globally | #1 open-source model Also #1 among Chinese models, with Xiaomi

Aligning with Your Own Voice: Self-Corrected Preference Learning for Hallucination Mitigation in LVLMs

arXiv:2604.24395v1 Announce Type: new Abstract: Large Vision-Language Models (LVLMs) frequently suffer from hallucinations. Existing preference learning-based approaches largely rely on proprietary mo

safetyarxiv-cs-ai

Agents

amazing! it’s like talking to an ai from the past

amazing! it’s like talking to an ai from the past Announcing Talkie: a new, open-weight historical LLM! We trained and finetuned a 13B model on a newly-curated dataset of only pre-1930 data. Try it be

agentsyohei-nakajima--x

Analytica: Soft Propositional Reasoning for Robust and Scalable LLM-Driven Analysis

arXiv:2604.23072v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly tasked with complex real-world analysis (e.g., in financial forecasting, scientific discovery), yet t

safetyarxiv-cs-ai

AsyncShield: A Plug-and-Play Edge Adapter for Asynchronous Cloud-based VLA Navigation

arXiv:2604.24086v1 Announce Type: cross Abstract: While Vision-Language-Action (VLA) models have been demonstrated possessing strong zero-shot generalization for robot control, their massive parameter

Local Ai

Benchmarking Source-Sensitive Reasoning in Turkish: Humans and LLMs under Evidential Trust Manipulation

arXiv:2604.24665v1 Announce Type: cross Abstract: This paper investigates whether source trustworthiness shapes Turkish evidential morphology and whether large language models (LLMs) track this sensit

local-aiarxiv-cs-ai

Beyond Local vs. External: A Game-Theoretic Framework for Trustworthy Knowledge Acquisition

arXiv:2604.23413v1 Announce Type: new Abstract: Cloud-hosted Large Language Models (LLMs) offer unmatched reasoning capabilities and dynamic knowledge, yet submitting raw queries to these external ser

BIR-Adapter: A parameter-efficient diffusion adapter for blind image restoration

arXiv:2509.06904v3 Announce Type: replace Abstract: We introduce the BIR-Adapter, a parameter-efficient diffusion adapter for blind image restoration. Diffusion-based restoration methods have demonstr

ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents

arXiv:2604.23781v1 Announce Type: new Abstract: Language-model agents are increasingly used as persistent coworkers that assist users across multiple working days. During such workflows, the surroundi

Computational Design and Co-Robotic Fabrication for Material Reuse in Architecture

arXiv:2604.24648v1 Announce Type: new Abstract: Climate change and resource depletion demand a shift from the dominant linear 'take-make-use-dispose' paradigm of construction toward circular, low-wast

applicationsarxiv-cs-ro

Continual Calibration: Coverage Can Collapse Before Accuracy in Lifelong LLM Fine-Tuning

arXiv:2604.23987v1 Announce Type: new Abstract: Continual learning for large language models is typically evaluated through accuracy retention under sequential fine-tuning. We argue that this perspect

researcharxiv-cs-lg

Cortex-Inspired Continual Learning: Unsupervised Instantiation and Recovery of Functional Task Networks

arXiv:2604.24637v1 Announce Type: cross Abstract: Block-sequential continual learning demands that a single model both protect prior solutions from catastrophic forgetting and efficiently infer at inf

Coverage-Based Calibration for Post-Training Quantization via Weighted Set Cover over Outlier Channels

arXiv:2604.24008v1 Announce Type: new Abstract: Post-Training Quantization (PTQ) compresses large language models to low bit-widths using a small calibration set, and its quality depends strongly on w