All entries

Internally Referenced Low-Light Enhancement

model-releasesarxiv-cs-cv

arXiv:2605.28605v1 Announce Type: new Abstract: Self-supervised low-light image enhancement (LLIE) is highly appealing as it eliminates the reliance on external paired data. However, the lack of exter

Interpretability-Guided Layer Selection over Subspace Projection: SAEs as Stethoscopes, Not Scalpels, for Raw Task Vector Model Editing

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DGX agent

arXiv:2605.28649v1 Announce Type: cross Abstract: LLMs increasingly require surgical model editing to enhance domain-specific capabilities without incurring the computational cost or catastrophic forg

model-releasesarxiv-cs-cl

28 May 2026

Intra-YOLO: A Small Object Detection Model for Caries and Molar-Incisor Hypomineralization in Intraoral Photography Based on Transfer Learning with Reinforcement Learning

arXiv:2605.28157v1 Announce Type: new Abstract: This study developed a computer-aided diagnosis (CAD) system for detecting caries and molar-incisor hypomineralization (MIH) in intraoral photographs. T

researcharxiv-cs-cv

Inversely Learning Transferable Rewards via Abstracted States

arXiv:2501.01669v4 Announce Type: replace Abstract: Inverse reinforcement learning (IRL) has progressed significantly toward accurately learning the underlying rewards in both discrete and continuous

IPO-Mine: A Toolkit and Dataset for Section-Structured Analysis of Long, Multimodal IPO Documents

arXiv:2605.28714v1 Announce Type: cross Abstract: An Initial Public Offering (IPO) filing is a document released when a private firm goes public, allowing individual (retail) investors to purchase its

IRDS: Interpretable RLVR Data Selection via Verifier-Coupled Sparse Autoencoder Coverage

arXiv:2605.28247v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a key technique for en- hancing LLM reasoning, yet its data ineffi- ciency remains a

IRPO: Boosting Image Restoration via Post-training GRPO

arXiv:2512.00814v3 Announce Type: replace Abstract: Post-training has become effective for high-level generation, but its role in low-level vision remains underexplored. Existing image restoration met

researcharxiv-cs-cv

Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency

arXiv:2605.27946v1 Announce Type: cross Abstract: Backpropagation is the default learning rule for artificial neural networks and is often treated as the settled approach whenever differentiability is

Isometry pursuit

arXiv:2411.18502v2 Announce Type: replace-cross Abstract: Isometry pursuit is a convex algorithm for identifying orthonormal column-submatrices of wide matrices. It consists of a novel normalization m

Iterative Causal Discovery: Per-Edge Impossibility Certificates, Tier-Aware Oracle Queries, and the 1+K Lower Bound

arXiv:2605.27477v1 Announce Type: cross Abstract: Causal-discovery algorithms return a directed graph, yet provide no principled means of distinguishing edge directions identified by the data from tho

Janus-LoRA: A Balanced Low-Rank Adaptation for Continual Learning

model-releasesarxiv-cs-cv

arXiv:2605.28495v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) has emerged as a promising paradigm for Continual Learning. It independently updates its low-rank factors (A and B), creating

JECA^2: Judgment-Explanation Consistent Adversarial Attack against Forensic Vision-Language Models

arXiv:2605.28609v1 Announce Type: new Abstract: Forensic vision-language models (VLMs) have recently been developed to detect image tampering and provide natural-language explanations. However, their

safetyarxiv-cs-cv

JMedEthicBench: A Multi-Turn Conversational Benchmark for Evaluating Medical Safety in Japanese Large Language Models

arXiv:2601.01627v3 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) are increasingly deployed in healthcare field, it becomes essential to carefully evaluate their medical safety

Joint Training of Multi-Token Prediction in Reinforcement Learning via Optimal Coefficient Calibration

arXiv:2605.28184v1 Announce Type: new Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as the standard paradigm for improving reasoning capability of large language models,

safetyarxiv-cs-lg

Keyphrase Generative Representation of Youth Crisis Conversations Beyond Static Taxonomies

arXiv:2605.27546v1 Announce Type: new Abstract: Crisis Responders (CRs) rapidly assess thousands of youth SMS conversations each year to identify mental health concerns and guide support. Yet youth di

tutorialsarxiv-cs-cl

Knowing When to Ask: Segment-Level Credit Assignment for LLM Tool Use

arXiv:2605.27788v1 Announce Type: cross Abstract: Humans know when to reach for help e.g. 347 imes 28 warrants a calculator while 2+2 does not. Language models do not. Prompt-based approaches can inst

agentsarxiv-cs-cl

Knowledge Dependency Estimation for Reliable Question Answering

arXiv:2605.28047v1 Announce Type: new Abstract: Reliable question answering requires identifying not only whether an answer is correct, but also which available knowledge the prediction depends on. In

researcharxiv-cs-cl

KSAFE-MM: A Multimodal Safety Benchmark via Localized Contextualization for Korean Cultural Risks

model-releasesarxiv-cs-cl

arXiv:2605.28013v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) exacerbate safety risks by introducing vulnerabilities across multiple modalities, such as language and vision.

KVoiceBench, KOpenAudioBench, and KMMAU: Agent-Driven Korean Speech Benchmarks for Evaluating SpeechLMs

arXiv:2605.27984v1 Announce Type: cross Abstract: Speech language models (SpeechLMs) have achieved substantial progress by extending large language models (LLMs) to the speech modality. However, Speec

La-Proteina: Atomistic Protein Generation via Partially Latent Flow Matching

arXiv:2507.09466v2 Announce Type: replace Abstract: Recently, many generative models for de novo protein structure design have emerged. Yet, only few tackle the difficult task of directly generating f

LACUNA: Safe Agents as Recursive Program Holes

arXiv:2605.28617v1 Announce Type: new Abstract: LLM agents increasingly act by writing code, yet a split persists between the runtime that drives the agent and the code the model writes. The runtime o

safetyarxiv-cs-ai

Laguna M.1/XS.2 Technical Report

arXiv:2605.27605v1 Announce Type: new Abstract: We present Laguna M.1 and Laguna XS.2, two Mixture-of-Experts foundation models built for long-horizon, agentic coding: M.1 has 225.8B total parameters

LaneRoPE: Positional Encoding for Collaborative Parallel Reasoning and Generation

arXiv:2605.27570v1 Announce Type: new Abstract: Parallel LLM test-time scaling techniques (e.g., best-of-N) require drawing N>1 sequences conditioned on the same input prompt. These methods boost accu

Large Language Models Approach Expert Pedagogical Quality in Math Tutoring but Differ in Instructional and Linguistic Profiles

arXiv:2512.20780v3 Announce Type: replace Abstract: Recent work has explored the use of large language models (LLMs) to generate tutoring responses in mathematics, yet it remains unclear how closely t

agentsarxiv-cs-cl

Large Language Models as Automatic Annotators and Annotation Adjudicators for Fine-Grained Opinion Analysis

arXiv:2601.16800v3 Announce Type: replace Abstract: Fine-grained opinion analysis of text provides a detailed understanding of expressed sentiments, including the addressed entity. Although this level

agentsarxiv-cs-cl

Latent-Conditioned Parameterized Quantum Circuits as Universal Approximators for Distributions over Quantum States

arXiv:2605.28690v1 Announce Type: cross Abstract: Many applications in quantum simulation, quantum chemistry, and quantum machine learning require not a single quantum state but an ensemble of states

Latent Diffusion for Missing Data

arXiv:2605.28427v1 Announce Type: new Abstract: Diffusion models have emerged as powerful generative approaches for missing-data imputation, yet most existing methods operate directly in data space an

Law of Neural Interaction: Depth-Width Shape, Interaction Efficiency, and Generalization

model-releasesarxiv-cs-lg

arXiv:2605.27989v1 Announce Type: new Abstract: The guidance of scaling laws has increased the resource demands of modern large language models (LLMs), yet it remains questionable whether these models

LCO: LLM-based Constraint Optimization for Safer Agentic LLMs in Real-world Tasks

model-releasesarxiv-cs-cl

arXiv:2605.27375v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly acting as autonomous agents, but their continuous interaction with the environment can lead to in-context

Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents

arXiv:2605.28775v1 Announce Type: cross Abstract: Computer-use agents (CUAs) have recently made substantial progress, but deploying a separate large expert for each software domain remains expensive.

agentsarxiv-cs-ai

Learn from your own latents and not from tokens: A sample-complexity theory

arXiv:2605.27734v1 Announce Type: new Abstract: Generative models, from diffusion models to large language models, achieve remarkable performance but at a cost in training data orders of magnitude lar

Learning a Kinodynamic Trajectory Manifold for Impact-Aware Compliant Catching of Fast-Moving Objects

arXiv:2605.28462v1 Announce Type: new Abstract: Fast catching of free-flying objects is difficult because of short reaction time, impact uncertainty, and kinodynamic constraints. We use reinforcement

tutorialsarxiv-cs-ro

Applications

Learning after COVID-19 and the ICT career aspirations: Are students entering the AI era with weaker skills?

arXiv:2605.27391v1 Announce Type: cross Abstract: This paper examines whether students are entering the generative AI era with sufficiently strong educational foundations, focusing on the relationship

applicationsarxiv-cs-ai

Learning Compositional Latent Structure with Vector Networks

arXiv:2605.28007v1 Announce Type: cross Abstract: Deep networks are powerful function approximators, but they typically store many different computations in shared weight matrices, making it difficult

Learning Correlated Reward Models: Statistical Barriers and Opportunities

arXiv:2510.15839v2 Announce Type: replace Abstract: Random Utility Models (RUMs) are a classical framework for modeling user preferences and play a key role in reward modeling for Reinforcement Learni

Learning Deliberately, Acting Intuitively: Unlocking Test-Time Reasoning in Multimodal LLMs

arXiv:2507.06999v2 Announce Type: replace-cross Abstract: Reasoning is essential for large language models (LLMs), especially in complex tasks such as mathematical problem solving. However, multimodal

safetyarxiv-cs-cl

Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent

arXiv:2605.28612v1 Announce Type: new Abstract: Parity functions are fundamental Boolean operations with critical applications across machine learning, cryptography, and error correction. Yet, learnin

safetyarxiv-cs-lg

Learning Logical Operations for Arbitrary Quantum Error Correction Codes

arXiv:2605.28162v1 Announce Type: cross Abstract: Logical operations are essential for quantum computation within quantum error-correcting codes. However, discovering their physical realizations is ch

Learning Tangent Bundles and Characteristic Classes with Autoencoder Atlases

arXiv:2602.22873v2 Announce Type: replace-cross Abstract: We introduce a theoretical framework that connects multi-chart autoencoders in manifold learning with the classical theory of vector bundles a

Learning the Error Patterns of Language Models

arXiv:2605.28328v1 Announce Type: cross Abstract: When generating outputs for domains with specific validity constraints (e.g., a program should compile), LLMs often fail in a small number of focused

tutorialsarxiv-cs-ai

Learning Theory of the SVRG: Generalization and Convergence Analysis

arXiv:2605.28513v1 Announce Type: cross Abstract: Variance reduction (VR) methods employ stochastic gradients with decreasing variance, and they have been widely applied to solve large-scale optimizat

Learning to Assess the Reliability of Number-of-Runs Estimation in Stochastic Optimization

arXiv:2605.28309v1 Announce Type: new Abstract: In large-scale benchmarking of stochastic optimization algorithms, the key challenge is no longer whether repeated runs are needed for reliability, but

Learning to Assign Prediction Tasks to Agents with Capacity Constraints

arXiv:2605.27999v1 Announce Type: cross Abstract: We address the problem of learning to assign prediction tasks to one agent from a set of available human or AI agents. In particular, we focus on the

safetyarxiv-cs-ai

Learning to Bid in Repeated Second-Price Auctions with Dynamic Values and Aggregated Feedback

arXiv:2605.28133v1 Announce Type: new Abstract: We study the problem of learning to bid when the bidder's value is dynamic, i.e., when the current value depends on past outcomes. Specifically, we cons

safetyarxiv-cs-lg

Learning to Label: A Reinforced Self-Evolving Framework for Semi-supervised Referring Expression Segmentation

arXiv:2605.28239v1 Announce Type: new Abstract: Semi-supervised referring expression segmentation (SS-RES) aims to achieve precise pixel-level language grounding under limited annotation, yet suffers

researcharxiv-cs-cv

Applications

Learning to target with network interference

arXiv:2605.27794v1 Announce Type: cross Abstract: This paper studies adaptive targeting under network interference in a bandit setting, where treatments applied to one individual may affect others thr

applicationsarxiv-cs-lg

Learning to Translate from Soft to Hard LLM Prompts

model-releasesarxiv-cs-cl

arXiv:2605.27642v1 Announce Type: new Abstract: Soft prompt tuning is a parameter-efficient method for adapting LLMs to specific tasks, but suffers from a lack of interpretability. Building on recent

Learning What to Recommend: Minimax Optimal Simple Regret in Logistic Bandits

arXiv:2601.21167v2 Announce Type: replace Abstract: We study stochastic logistic bandits with d-dimensional action features under the simple-regret objective, where a learner uses T rounds of explorat