Stay updated with the latest research and technology news
arXiv:2502.08941v3 Announce Type: replace Abstract: This paper analyzes multi-step temporal difference (TD)-learning algorithms within the ``deadly triad'' scenario, characterized by linear function approximation, off-policy learning,...
arXiv:2502.09257v4 Announce Type: replace Abstract: We study the problem of contextual combinatorial semi-bandits, where input contexts are mapped into subsets of size $m$ of a...
arXiv:2502.10898v2 Announce Type: replace Abstract: When the game Lights Out is played according to an algorithm specifying the player's sequence of moves, it can be...
arXiv:2502.11034v2 Announce Type: replace Abstract: Loss spikes remain a persistent obstacle in large-scale language model pretraining. While previous research has attempted to identify the root...
arXiv:2502.13069v3 Announce Type: replace Abstract: AI agents are increasingly being deployed to automate tasks, often based on underspecified user instructions. Making unwarranted assumptions to compensate...
arXiv:2503.04940v2 Announce Type: replace Abstract: Emergent Language (EL) focuses on the emergence of communication among artificial agents. Although symbolic communication channels more closely mirror the...
arXiv:2503.06084v2 Announce Type: replace Abstract: Visual prompt tuning offers significant advantages for adapting pre-trained visual foundation models to specific tasks. However, current research provides limited...
arXiv:2503.07853v2 Announce Type: replace Abstract: Traditional classifiers treat all labels as mutually independent, thereby considering all negative classes to be equally incorrect. This approach fails...
arXiv:2503.08049v3 Announce Type: replace Abstract: The reliance on Deep Neural Network (DNN)-based classifiers in safety-critical and real-world applications necessitates Open-Set Recognition (OSR). OSR enables the...
arXiv:2503.11842v2 Announce Type: replace Abstract: Test-time training (TTT) methods explicitly update the weights of a model to adapt to the specific test instance, and they...
arXiv:2503.12047v3 Announce Type: replace Abstract: Gait recognition has emerged as a robust biometric modality due to its non-intrusive nature. Conventional gait recognition methods mainly rely...
arXiv:2503.12670v3 Announce Type: replace Abstract: The construction of stable, conservative, and accurate volume dissipation is extended to discretizations that possess a generalized summation-by-parts (SBP) property...
arXiv:2503.13444v3 Announce Type: replace Abstract: Videos, with their unique temporal dimension, demand precise grounded understanding, where answers are directly linked to visual, interpretable evidence. Despite...
arXiv:2503.14637v2 Announce Type: replace Abstract: How do humans move? Advances in reinforcement learning (RL) have produced impressive results in capturing human motion using physics-based humanoid...
arXiv:2503.14720v2 Announce Type: replace Abstract: We present ShapeShift, a method for arranging rigid objects into configurations that visually convey semantic concepts specified by natural language....
arXiv:2503.15937v5 Announce Type: replace Abstract: We propose V-Droid, a mobile GUI task automation agent. Unlike previous mobile agents that utilize Large Language Models (LLMs) as...
arXiv:2503.16783v3 Announce Type: replace Abstract: The security of many Proof-of-Stake (PoS) payment systems relies on quorum-based State Machine Replication (SMR) protocols. While classical analyses assume...
arXiv:2503.19356v3 Announce Type: replace Abstract: AI models have made significant strides in recent years in their ability to describe and answer questions about real-world images....
arXiv:2503.21258v2 Announce Type: replace Abstract: Few-shot class-incremental Learning (FSCIL) enables models to learn new classes from limited data while retaining performance on previously learned classes....
arXiv:2503.23377v2 Announce Type: replace Abstract: This paper introduces JavisDiT, a novel Joint Audio-Video Diffusion Transformer designed for synchronized audio-video generation (JAVG). Based on the powerful...
arXiv:2503.23608v2 Announce Type: replace Abstract: We model human and animal learning by computing with high-dimensional vectors (H = 10,000 for example). The architecture resembles traditional...
arXiv:2503.24298v2 Announce Type: replace Abstract: Fine-grained understanding of human actions is essential for safe and intuitive human--robot interaction. We study the challenge of recognizing nearly...
arXiv:2504.02996v2 Announce Type: replace Abstract: Methods addressing Learning with Noisy Labels (LNL) and multi-source Domain Generalization (DG) use training techniques to improve downstream task performance...
arXiv:2504.03349v2 Announce Type: replace Abstract: Recent advances in text recognition led to a paradigm shift for page-level recognition, from multi-step segmentation-based approaches to end-to-end attention-based...