12
ProsMAE: Multi-Source MAE Pretraining for ISUP Grade Classification
ProsMAE is a multi-source masked autoencoder framework designed to learn robust representations from gigapixel pathology slides for prostate cancer grading.
12
Feedback Manipulation Regularization: Enabling Offline Agent Alignment for Imitation Learning
Researchers have introduced a regularization method to improve offline agent alignment in imitation learning by leveraging human feedback and demonstrations.
12
Alignment Plausibility: A New Standard for Assuring AI in Healthcare
Researchers propose a new evaluation standard to address long-term risks and engagement-driven biases in healthcare-focused large language models.
12
MAVEN: A Multi-stage Agentic Annotation Pipeline for Video Reasoning Tasks
Researchers have introduced MAVEN, a multi-stage agentic pipeline designed to generate high-quality structured video annotations and reasoning traces for training vision-language models.
12
Aligning Clinical Needs and AI Capabilities: A Survey on LLMs for Medical Reasoning
A comprehensive survey evaluates the progress of large language models in medical reasoning by mapping clinical competencies to computational methods.
12
Context Graphs for Proactive Enterprise Agents
Researchers propose using a live relational data structure called a Context Graph to enable enterprise AI agents to proactively share information before being asked.
12
Graph-Regularized Deep Learning for EEG-Based Emotion Recognition with Psychologically-Grounded Label Structure
A new graph-regularized learning framework incorporates psychological interdependencies among emotion classes to improve EEG-based emotion recognition.
12
Adversarial Decoys: Misdirecting Attention-Based Defenses in ViT
The study introduces adversarial decoys, a method designed to bypass attention-based defenses in Vision Transformers by misdirecting token suppression mechanisms.
12
How Causal Abstraction Underpins Computational Explanation
This paper argues that causal abstraction theory provides a valuable framework for explaining how neural networks implement specific computational processes.
12
Robust Weighted Triangulation of Causal Effects Under Model Uncertainty
The paper proposes a robust weighted triangulation method to improve causal effect estimation when dealing with model uncertainty in observational data.
12
Second-Order Actor-Critic Methods for Discounted MDPs via Policy Hessian Decomposition
This paper introduces second-order actor-critic methods utilizing policy Hessian decomposition to accelerate convergence in discounted Markov decision processes.
12
Temporal Preference Concepts and their Functions in a Large Language Model
This work localizes the internal subgraph responsible for representing and resolving temporal preferences and trade-offs within a distilled large language model.
12
Generalization in offline RL: The structure is more important than the amount of pessimism
This paper demonstrates that optimal generalization in offline reinforcement learning depends on the structure of the pessimistic bias rather than the absolute level of conservatism.
12
MambaLIE: Scene Light Intensity-Boosted Low-Light Image Enhancement with State Space Model
Researchers have introduced MambaLIE, a state space model-based framework designed to enhance low-light images by leveraging scene light intensity.
12
Provably Optimal Learning Algorithms for Assistance Games
Researchers have introduced the first provably efficient learning algorithms for repeated assistance games to optimize shared reward functions between human and AI agents.
12
36 氪 AI 测评小程序上新！看真实体验与深度测评，多个新功能等你来玩~
Chinese tech media outlet 36Kr has updated its AI evaluation platform and WeChat mini-program to provide users with objective reviews and ratings of various AI products.
12
GeForce NOW Turns Up the Heat With New GeForce RTX 5080-Powered Toronto Server
NVIDIA has expanded its GeForce NOW cloud gaming service with a new server in Toronto powered by the GeForce RTX 5080 GPU.
12
China records most new unicorn start-ups in 5 years as AI and robotics boom
Driven by a surge in artificial intelligence and robotics investments, China added 67 new unicorn startups in the first half of 2026, marking its highest growth rate in nearly five years.
12
中信证券：港股扰动趋缓卖空料将收敛，短期推荐创新药、航空、机器人及工业金属
CITIC Securities recommends short-term investments in sectors including robotics, innovative drugs, and aviation as market short-selling pressures in Hong Kong are expected to ease.
12
MSRNet: A Multi-Scale Recursive Network for Camouflaged Object Detection
MSRNet is a multi-scale recursive network designed to improve the detection and segmentation of camouflaged objects in challenging visual environments.
12
Architecture Generalization with MetaNCA
The MetaNCA framework is introduced to improve the adaptability and architectural generalization of neural cellular automata models through localized interactions.
12
OmniFood-Bench: Evaluating VLMs for Nutrient Reasoning and Personalized Health Advice
OmniFood-Bench is a new benchmark designed to assess how well vision-language models can reason about nutritional content and provide personalized dietary advice.
12
Spectral Analysis of Dueling Q-Learning
This paper provides a theoretical spectral analysis of dueling Q-learning to better understand its convergence properties and performance in reinforcement learning.
12
MetaHGNIE: Meta-Path Induced Hypergraph Contrastive Learning in Heterogeneous Knowledge Graphs
The proposed MetaHGNIE framework utilizes hypergraph contrastive learning to better capture high-order interactions and estimate node importance in heterogeneous knowledge graphs.