model releases

Can LLMs Compress (and Decompress)? Evaluating Code Understanding and Execution via Invertibility

arXiv:2601.13398v2 Announce Type: replace Abstract: LLMs demonstrate strong performance on code benchmarks, yet consistent reasoning across forward and backward execution remains elusive. We present R

Model Releases

CASE: An Agentic AI Framework for Enhancing Scam Intelligence in Digital Payments

DGX agent

arXiv:2508.19932v2 Announce Type: replace Abstract: The proliferation of digital payment platforms has transformed commerce, offering unmatched convenience and accessibility globally. However, this gr

model-releasesarxiv-cs-ai

5 May 2026

Causal2Vec: Improving Decoder-only LLMs as Embedding Models through a Contextual Token

arXiv:2507.23386v3 Announce Type: replace Abstract: Decoder-only large language models (LLMs) have been increasingly adopted to build embedding models for diverse tasks. To overcome the inherent limit

CEZSAR: A Contrastive Embedding Method for Zero-Shot Action Recognition

arXiv:2605.01165v1 Announce Type: new Abstract: This paper proposes a novel Zero-Shot Action Recognition~(ZSAR) method based on contrastive learning. In ZSAR, we aim to classify examples from classes

CGFformer: Cluster-Guidance Frequency Transformer for Pansharpening

arXiv:2605.01490v1 Announce Type: new Abstract: Pansharpening aims to generate high-resolution multispectral (HRMS) images by fusing low-resolution multispectral (LRMS) images with high-resolution pan

Channel-Level Relation to Attentive Aggregation with Neighborhood-Homogeneity Constraint for Point Cloud Analysis

arXiv:2605.02357v1 Announce Type: new Abstract: In 3D point cloud understanding, the core challenge lies in accurately capturing discriminative features within complex neighborhoods, which directly af

Chart-FR1: Visual Focus-Driven Fine-Grained Reasoning on Dense Charts

arXiv:2605.01882v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have shown considerable potential in chart understanding and reasoning tasks. However, they still struggle with

Chebyshev-Augmented One-Shot Transfer Learning for PINNs on Nonlinear Differential Equations

arXiv:2605.01634v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) offer a flexible paradigm for solving differential equations by embedding governing laws into the training obje

Checkerboard: A Simple, Effective, Efficient and Learning-free Clean Label Backdoor Attack with Low Poisoning Budget

arXiv:2605.01298v1 Announce Type: cross Abstract: Backdoor attacks threaten the deep learning supply chain by poisoning a small fraction of the training data so that a model behaves normally on clean

CLaC at SemEval-2026 Task 6: Response Clarity Detection in Political Discourse

arXiv:2605.02170v1 Announce Type: new Abstract: In this paper, we present our system for SemEval-2026 Task 6 (CLARITY) on response clarity and evasion detection in question-answer pairs from U.S. pres

Class-Aware Adaptive Differential Privacy in Deep Learning for Sensor-Based Fall Detection

arXiv:2605.01679v1 Announce Type: cross Abstract: Fall detection is a critical task in healthcare, particularly for elderly people. Timely fall detection and treatment can prevent severe injuries. Sen

Cloud Engineer’s AI Toolkit: Sign up Now for a Developer Workshop Near You!

model-releasesgoogle-cloud-ai

The world of AI is rapidly shifting from experimental Large Language Models to an era of Agentic AI. In the agentic era, autonomous software agents act on behalf of employees and consumers—driving a f

CNN-based Multi-In-Multi-Out Model for Efficient Spatiotemporal Prediction

arXiv:2605.01277v1 Announce Type: new Abstract: Recently, Convolutional Neural Network (CNN) or Transformer architecture based models have been proposed to overcome the limitations of Recurrent Neural

CoAction: Cross-task Correlation-aware Pareto Set Learning

arXiv:2605.01712v1 Announce Type: new Abstract: Pareto set learning (PSL) is an emerging paradigm in multi-objective optimization that trains neural networks to map preference vectors to Pareto optima

COCORELI: Enforcing Execution Preconditions for Reliable Collaborative Instruction Following

arXiv:2509.04470v2 Announce Type: replace Abstract: Autonomous agents executing human instructions must operate reliably even when instructions are incomplete. While recent approaches improve detectio

CombinationTS: A Modular Framework for Understanding Time-Series Forecasting Models

arXiv:2605.01231v1 Announce Type: new Abstract: Recent progress in time-series forecasting has led to rapidly increasing architectural complexity, yet many reported State-of-the-Art gains are statisti

Component-Aware Self-Speculative Decoding in Hybrid Language Models

arXiv:2605.01106v1 Announce Type: new Abstract: Speculative decoding accelerates autoregressive inference by drafting candidate tokens with a fast model and verifying them in parallel with the target.

Compute Optimal Tokenization

arXiv:2605.01188v1 Announce Type: new Abstract: Scaling laws enable the optimal selection of data amount and language model size, yet the impact of the data unit, the token, on this relationship remai

Concepts Whisper While Syntax Shouts: Spectral Anti-Concentration and the Dual Geometry of Transformer Representations

arXiv:2605.01609v1 Announce Type: new Abstract: We test whether the causal inner product of itet{park2024linear} -- defined by the unembedding covariance Sigma -- enables cross-lingual concept transpo

Congestion-Aware Dynamic Axonal Delay for Spiking Neural Networks

arXiv:2605.01291v1 Announce Type: new Abstract: Spiking Neural Networks (SNNs) are widely regarded as an energy-efficient paradigm for modeling and processing temporal and event-driven information. In

Constructing Interpretable Features from Compositional Neuron Groups

arXiv:2506.10920v2 Announce Type: replace Abstract: A central goal for mechanistic interpretability has been to identify the right units of analysis in large language models (LLMs) that causally expla

ContextualJailbreak: Evolutionary Red-Teaming via Simulated Conversational Priming

arXiv:2605.02647v1 Announce Type: new Abstract: Large language models (LLMs) remain vulnerable to jailbreak attacks that bypass safety alignment and elicit harmful responses. A growing body of work sh

Control Reinforcement Learning: Interpretable Token-Level Steering of LLMs via Sparse Autoencoder Features

arXiv:2602.10437v3 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) decompose language model activations into interpretable features, but existing methods reveal only which features a

Controllable Logical Hypothesis Generation for Abductive Reasoning in Knowledge Graphs

model-releasesarxiv-cs-ai

arXiv:2505.20948v3 Announce Type: replace Abstract: Abductive reasoning in knowledge graphs aims to generate plausible logical hypotheses from observed entities, with broad applications in areas such

Coopetition-Gym v1: A Formally Grounded Platform for Mixed-Motive Multi-Agent Reinforcement Learning under Strategic Coopetition

arXiv:2605.02063v1 Announce Type: cross Abstract: We present Coopetition-Gym v1, a benchmark platform for mixed-motive multi-agent reinforcement learning under strategic coopetition. The platform comp

CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features

arXiv:2508.12535v3 Announce Type: replace Abstract: Sparse Autoencoders (SAEs) can extract interpretable features from large language models (LLMs) without supervision. However, their effectiveness in

CoSpaDi: Compressing LLMs via Calibration-Guided Sparse Dictionary Learning

arXiv:2509.22075v5 Announce Type: replace Abstract: Post-training compression of large language models (LLMs) often relies on low-rank weight approximations that represent each column of the weight ma

CP-SynC: Multi-Agent Zero-Shot Constraint Modeling in MiniZinc with Synthesized Checkers

arXiv:2605.01675v1 Announce Type: cross Abstract: Constraint Programming (CP) is a powerful paradigm for solving combinatorial problems, yet translating natural language problem descriptions into exec

Creating and Evaluating Figurative Language Dataset for Sindhi

arXiv:2605.01323v1 Announce Type: new Abstract: In this article, we introduce SiNFluD, a novel benchmark dataset for Sindhi figurative language classification. We first collect raw text from various b

Cross-Language Learning within Arabic Script for Low-Resource HTR

arXiv:2605.02089v1 Announce Type: new Abstract: Handwritten Text Recognition (HTR) under limited labeled data remains a challenging problem, particularly for Arabic-script languages. Although modern s

CyclicJudge: Mitigating Judge Bias Efficiently in LLM-based Evaluation

arXiv:2603.01865v3 Announce Type: replace Abstract: LLM-as-judge evaluation has become standard practice for open-ended model assessment; however, judges exhibit systematic biases that cannot be avera

datasette-referrer-policy 0.1

model-releasessimon-willison

Release: datasette-referrer-policy 0.1 The OpenStreetMap tiles on the Datasette global-power-plants demo weren't displaying correctly. This turned out to be caused by two bugs. The first is that the C

🚀 Day-0 MTP support for Gemma4 now available at vLLM with ready-to-use docker image! ⚡️Enjoy up to 3x faster decoding performance to superc…

model-releasesjeremy-howard--x

🚀 Day-0 MTP support for Gemma4 now available at vLLM with ready-to-use docker image! ⚡️Enjoy up to 3x faster decoding performance to supercharge your development with zero quality degradation! Check o

DBLP: Phase-Aware Bounded-Loss Transport for Burst-Resilient Distributed ML Training

arXiv:2605.01989v1 Announce Type: new Abstract: Distributed machine learning (ML) training has become a necessity with the prevalence of billion to trillion-parameter-scale models. While prior work ha

Decoding-Time Debiasing via Process Reward Models: From Controlled Fill-in to Open-Ended Generation

arXiv:2605.02348v1 Announce Type: new Abstract: Large language models pick up social biases from the data they are trained on and carry those biases into downstream applications, often reinforcing ste

Decompose and Recompose: Reasoning New Skills from Existing Abilities for Cross-Task Robotic Manipulation

arXiv:2605.01448v1 Announce Type: cross Abstract: Cross-task generalization is a core challenge in open-world robotic manipulation, and the key lies in extracting transferable manipulation knowledge f

Deep neural networks with Fisher vector encoding for medical image classification

arXiv:2605.01667v1 Announce Type: new Abstract: Orderless encoding methods have shown to improve Convolutional Neural Networks (CNNs) for image classification in the context of limited availability of

Deep Time Series Models: A Comprehensive Survey and Benchmark

arXiv:2407.13278v3 Announce Type: replace Abstract: Time series, characterized by a sequence of data points organized in a discrete-time order, are ubiquitous in real-world scenarios. Unlike other dat

DeepSeek v4 works fine, but it’s not the frontier-pressing moment we saw with Kimi 2.6. On Notion eval data, it’s similar performance to GPT…

model-releaseskimi-moonshot--x

DeepSeek v4 works fine, but it’s not the frontier-pressing moment we saw with Kimi 2.6. On Notion eval data, it’s similar performance to GPT 5.2, with understandable failings. Most interesting — it do

Democratizing the medieval English legal tradition

arXiv:2605.00977v1 Announce Type: cross Abstract: The record of the beginning of the most widespread legal system in the world is contained in millions of pages of handwritten text. Most of the record

Developing a Strong Pre-Trained Base Model for Plant Leaf Disease Classification

arXiv:2605.01283v1 Announce Type: new Abstract: Plants, crops and their yields are essential to our very existence, but diseases and pests cause large losses every year. As such it is vital to ensure

Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score

arXiv:2603.23985v2 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated remarkable capabilities, but their massive scale poses significant challenges for practical deploymen

Differentiable Multiphysics Co-Optimization via Implicit Neural Representations: A Transient Hamburger-Cooking Benchmark

arXiv:2605.01040v1 Announce Type: cross Abstract: The co-optimization of geometry and physical parameters remains challenging in transient multiphysics systems involving moving boundaries, nonlinear m

Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models

arXiv:2602.00217v2 Announce Type: replace Abstract: Large language models (LLMs) achieve remarkable performance through ever-increasing parameter counts, but scaling incurs steep computational costs.

Doc and sources: Google is testing an agent in the Gemini app, internally named Remy, that can integrate with Google services to take actions on a user's behalf (Hugh Langley/Business Insider)

Hugh Langley / Business Insider: Doc and sources: Google is testing an agent in the Gemini app, internally named Remy, that can integrate with Google services to take actions on a user's behalf — - Go

model-releasestechmeme

Dual-branch Robust Unlearnable Examples

arXiv:2605.01718v1 Announce Type: new Abstract: Unlearnable examples (UEs) aim to compromise model training by injecting imperceptible perturbations to clean samples. However, existing UE schemes exhi

DuFal: Dual-Frequency-Aware Learning for High-Fidelity Extremely Sparse-view CBCT Reconstruction

arXiv:2601.15416v2 Announce Type: replace Abstract: Sparse-view Cone-Beam Computed Tomography reconstruction from limited X-ray projections remains a challenging problem in medical imaging due to the

DurableUn: Quantization-Induced Recovery Attacks in Machine Unlearning

arXiv:2605.02196v1 Announce Type: new Abstract: Machine unlearning aims to remove specified training data to satisfy privacy regulations such as GDPR. However, existing evaluations assume identical pr