research

Interpretable Deep Learning for Stock Returns: A Consensus-Bottleneck Asset Pricing Model

arXiv:2512.16251v5 Announce Type: replace-cross Abstract: We introduce the Consensus-Bottleneck Asset Pricing Model (CB-APM), which embeds aggregate analyst consensus as a structural bottleneck, treat

Research

Introducing Background Temperature to Characterise Hidden Randomness in Large Language Models

DGX agent

arXiv:2604.22411v1 Announce Type: new Abstract: Even when decoding with temperature T=0, large language models (LLMs) can produce divergent outputs for identical inputs. Recent work by Thinking Machin

researcharxiv-cs-ai

27 Apr 2026

Iterative Model-Learning Scheme via Gaussian Processes for Nonlinear Model Predictive Control of (Semi-)Batch Processes

arXiv:2604.22672v1 Announce Type: new Abstract: Batch processes are inherently transient and typically nonlinear, motivating nonlinear model predictive control (NMPC). However, adopting NMPC is hinder

Joint Embedding Variational Bayes

arXiv:2602.05639v4 Announce Type: replace Abstract: We introduce Variational Joint Embedding (VJE), a reconstruction-free latent-variable framework for non-contrastive self-supervised learning in repr

Large Language Models Decide Early and Explain Later

arXiv:2604.22266v1 Announce Type: new Abstract: Large Language Models often achieve strong performance by generating long intermediate chain-of-thought reasoning. However, it remains unclear when a mo

LATMiX: Learnable Affine Transformations for Microscaling Quantization of LLMs

arXiv:2602.17681v2 Announce Type: replace-cross Abstract: Post-training quantization (PTQ) is a widely used approach for reducing the memory and compute costs of large language models (LLMs). Recent s

LayerBoost: Layer-Aware Attention Reduction for Efficient LLMs

arXiv:2604.22050v1 Announce Type: cross Abstract: Transformers are mostly relying on softmax attention, which introduces quadratic complexity with respect to sequence length and remains a major bottle

Lifting Unlabeled Internet-level Data for 3D Scene Understanding

arXiv:2604.01907v2 Announce Type: replace-cross Abstract: Annotated 3D scene data is scarce and expensive to acquire, while abundant unlabeled videos are readily available on the internet. In this pap

LLM+Graph@VLDB'2025 Workshop Summary

arXiv:2604.02861v2 Announce Type: replace-cross Abstract: The integration of large language models (LLMs) with graph-structured data has become a pivotal and fast evolving research frontier, drawing s

Measuring and Mitigating Persona Distortions from AI Writing Assistance

arXiv:2604.22503v1 Announce Type: new Abstract: Hundreds of millions of people use artificial intelligence (AI) for writing assistance. Here, we evaluated how AI writing assistance distorts writer per

Mechanistic Interpretability of Antibody Language Models Using SAEs

arXiv:2512.05794v2 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within la

Mixed Membership sub-Gaussian Models

arXiv:2604.22633v1 Announce Type: cross Abstract: The Gaussian mixture model is widely used in unsupervised learning, owing to its simplicity and interpretability. However, a fundamental limitation of

Model Predictive Control of Hybrid Dynamical Systems

arXiv:2604.21989v1 Announce Type: cross Abstract: The problem of controlling hybrid dynamical systems using model predictive control (MPC) is formulated and sufficient conditions for asymptotic stabil

Multi-Token Prediction via Self-Distillation

arXiv:2602.06019v2 Announce Type: replace Abstract: Existing techniques for accelerating language model inference, such as speculative decoding, require training auxiliary speculator models and buildi

MultiTok: Variable-Length Tokenization for Efficient LLMs Adapted from LZW Compression

arXiv:2410.21548v3 Announce Type: replace Abstract: Large language models have drastically changed the prospects of AI by introducing technologies for more complex natural language processing. However

Neural Recovery of Historical Lexical Structure in Bantu Languages from Modern Data

arXiv:2604.22730v1 Announce Type: cross Abstract: We investigate whether neural models trained exclusively on modern morphological data can recover cross-lingual lexical structure consistent with hist

NeuronMLP: Efficient LLM Inference via Singular Value Decomposition Compression and Tiling on AWS Trainium

arXiv:2510.25977v4 Announce Type: replace Abstract: Emerging AI accelerators have started to gain attention and offer new opportunities for efficient inference of large language models (LLMs). Trainiu

New work with @AlecRad and @DavidDuvenaud: Have you ever dreamed of talking to someone from the past? Introducing talkie, a 13B model traine…

New work with @AlecRad and @DavidDuvenaud: Have you ever dreamed of talking to someone from the past? Introducing talkie, a 13B model trained only on pre-1931 text. Vintage models should help us to un

researchkarpathy--x

NiuTrans.LMT: Toward Inclusive and Scalable Multilingual Machine Translation with LLMs

arXiv:2511.07003v2 Announce Type: replace Abstract: Large language models have significantly advanced Multilingual Machine Translation (MMT), yet scaling to many languages while keeping quality robust

Non-Minimal Sampling and Consensus for Prohibitively Large Datasets

arXiv:2604.22518v1 Announce Type: new Abstract: We introduce NONSAC (Non-Minimal Sampling and Consensus), a general framework for robust and scalable model estimation from arbitrarily large datasets c

NRGS: Neural Regularization for Robust 3D Semantic Gaussian Splatting

arXiv:2604.22439v1 Announce Type: new Abstract: We propose a neural regularization method that refines the noisy 3D semantic field produced by lifting multi-view inconsistent 2D features, in order to

Null-Space Flow Matching for MIMO Channel Estimation in Latency-Constrained Systems

arXiv:2604.22005v1 Announce Type: cross Abstract: Accurate yet low-latency channel state information (CSI) acquisition is essential for multiple-input multiple-output (MIMO) communication systems. Whi

OmniOVCD: Streamlining Open-Vocabulary Change Detection with SAM 3

arXiv:2601.13895v2 Announce Type: replace-cross Abstract: Change Detection (CD) is a fundamental task in remote sensing. It monitors the evolution of land cover over time. Based on this, Open-Vocabula

On Pareto Optimality for Parametric Choice Bandits

arXiv:2501.19277v4 Announce Type: replace-cross Abstract: We study online assortment optimization under stochastic choice when a decision maker simultaneously values cumulative revenue performance and

Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe

arXiv:2604.22676v1 Announce Type: new Abstract: Graph neural networks achieve strong node-classification accuracy, but their learned message passing entangles ego attributes, neighborhood smoothing, h

Optimal Lower Bounds for Online Multicalibration

arXiv:2601.05245v2 Announce Type: replace Abstract: We prove tight lower bounds for online multicalibration, establishing an information-theoretic separation from marginal calibration. In the general

OREN: Octree Residual Network for Real-Time Euclidean Signed Distance Mapping

arXiv:2510.18999v2 Announce Type: replace-cross Abstract: Reconstructing signed distance functions (SDFs) from point cloud data benefits many robot autonomy capabilities, including localization, mappi

Outcome Rewards Do Not Guarantee Verifiable or Causally Important Reasoning

arXiv:2604.22074v1 Announce Type: new Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) on chain-of-thought reasoning has become a standard part of language model post-training recipes.

Pack only the essentials: Adaptive dictionary learning for kernel ridge regression

arXiv:2604.22386v1 Announce Type: cross Abstract: One of the major limits of kernel ridge regression (KRR) is that storing and manipulating the kernel matrix K_n for n samples requires O(n^2) space, w

PAGaS: Pixel-Aligned 1DoF Gaussian Splatting for Depth Refinement

arXiv:2604.22129v1 Announce Type: new Abstract: Gaussian Splatting (GS) has emerged as an efficient approach for high-quality novel view synthesis. While early GS variants struggled to accurately mode

Performance Anomaly Detection in Athletics: A Benchmarking System with Visual Analytics

arXiv:2604.21953v1 Announce Type: new Abstract: Anti-doping programs rely on biological testing to detect performance-enhancing drugs, but such testing costs over $800 per sample and is limited by sho

Point & Grasp: Flexible Selection of Out-of-Reach Objects Through Probabilistic Cue Integration

arXiv:2604.22491v1 Announce Type: cross Abstract: Selecting out-of-reach objects is a fundamental task in mixed reality (MR). Existing methods rely on a single cue or deterministically fuse multiple c

researcharxiv-cs-ro

PoLO: Proof-of-Learning and Proof-of-Ownership at Once with Chained Watermarking

arXiv:2505.12296v2 Announce Type: replace-cross Abstract: Our evaluation shows that PoLO achieves extbf{99%} watermark detection accuracy for ownership verification, while preserving data privacy and

Preference Heads in Large Language Models: A Mechanistic Framework for Interpretable Personalization

arXiv:2604.22345v1 Announce Type: new Abstract: Large Language Models (LLMs) exhibit strong implicit personalization ability, yet most existing approaches treat this behavior as a black box, relying o

PreMoE: Proactive Inference for Efficient Mixture-of-Experts

arXiv:2505.17639v3 Announce Type: replace Abstract: Mixture-of-Experts (MoE) models offer dynamic computation, but are typically deployed as static full-capacity models, missing opportunities for depl

Presenting DiaData for Research on Type 1 Diabetes

arXiv:2508.09160v2 Announce Type: replace Abstract: Type 1 diabetes (T1D) is an autoimmune disorder that leads to the destruction of insulin-producing cells, resulting in insulin deficiency, as to why

Privacy Leakage via Output Label Space and Differentially Private Continual Learning

arXiv:2411.04680v5 Announce Type: replace Abstract: Differential privacy (DP) is a formal privacy framework that enables training machine learning (ML) models while protecting individuals' data. As po

PrivUn: Unveiling Latent Ripple Effects and Shallow Forgetting in Privacy Unlearning

arXiv:2604.22076v1 Announce Type: cross Abstract: Large language models (LLMs) often memorize private information during training, raising serious privacy concerns. While machine unlearning has emerge

Protect the Brain When Treating the Heart: A Convolutional Neural Network for Detecting Emboli

arXiv:2604.22258v1 Announce Type: cross Abstract: Gaseous microemboli (GME) represent a common complication of cardiac structural interventions across both surgical and transcatheter approaches. Trans

Quality-Driven Selective Mutation for Deep Learning

arXiv:2604.22640v1 Announce Type: cross Abstract: Mutants support testing and debugging in two roles: (i) as test goals and (ii) as substitutes for real faults. Hard-to-kill mutants provide better gui

Read the release notes: https://github.com/keras-team/kinetic/releases/tag/0.0.2

researchfrancois-chollet--x

Kinetic version 0.0.2 release notes are available on GitHub, detailing updates and features for this Keras-related project. The announcement was shared by François Chollet, Keras creator, on X (former

ReLIC-SGG: Relation Lattice Completion for Open-Vocabulary Scene Graph Generation

arXiv:2604.22546v1 Announce Type: new Abstract: Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible relation phrases beyond a fixed predicate set. Existing method

Removing Sandbagging in LLMs by Training with Weak Supervision

arXiv:2604.22082v1 Announce Type: cross Abstract: As AI systems begin to automate complex tasks, supervision increasingly relies on weaker models or limited human oversight that cannot fully verify ou

Rethinking Math Reasoning Evaluation: A Robust LLM-as-a-Judge Framework Beyond Symbolic Rigidity

arXiv:2604.22597v1 Announce Type: new Abstract: Recent advancements in large language models have led to significant improvements across various tasks, including mathematical reasoning, which is used

Rethinking Token Pruning for Historical Screenshots in GUI Visual Agents: Semantic, Spatial, and Temporal Perspectives

arXiv:2603.26041v3 Announce Type: replace Abstract: In recent years, GUI visual agents built upon Multimodal Large Language Models (MLLMs) have demonstrated strong potential in navigation tasks. Howev

Revisiting Geometric Obfuscation with Dual Convergent Lines for Privacy-Preserving Image Queries in Visual Localization

arXiv:2604.22310v1 Announce Type: new Abstract: Privacy-Preserving Image Queries (PPIQ) are an emerging mechanism for cloud-based visual localization, enabling pose estimation from obfuscated features

Revisiting Neural Activation Coverage for Uncertainty Estimation

arXiv:2604.22360v1 Announce Type: new Abstract: Neural activation coverage (NAC) is a recently-proposed technique for out-of-distribution detection and generalization. We build upon this promising fou

RouteLMT: Learned Sample Routing for Hybrid LLM Translation Deployment

arXiv:2604.22520v1 Announce Type: new Abstract: Large Language Models (LLMs) have achieved remarkable performance in Machine Translation (MT), but deploying them at scale remains prohibitively expensi