Archives AI News

Variational Neural Networks for Observable Thermodynamics (V-NOTS)

arXiv:2509.09899v1 Announce Type: new Abstract: Much attention has recently been devoted to data-based computing of evolution of physical systems. In such approaches, information about data points from past trajectories in phase space is used to reconstruct the equations of motion…

MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs

arXiv:2506.07899v3 Announce Type: replace-cross Abstract: Language models deployed in real-world systems often require post-hoc updates to incorporate new or corrected knowledge. However, editing such models efficiently and reliably-without retraining or forgetting previous information-remains a major challenge. Existing methods for lifelong…

Latency and Token-Aware Test-Time Compute

arXiv:2509.09864v1 Announce Type: new Abstract: Inference-time scaling has emerged as a powerful way to improve large language model (LLM) performance by generating multiple candidate responses and selecting among them. However, existing work on dynamic allocation for test-time compute typically considers…

Your Image is Secretly the Last Frame of a Pseudo Video

arXiv:2410.20158v3 Announce Type: replace-cross Abstract: Diffusion models, which can be viewed as a special case of hierarchical variational autoencoders (HVAEs), have shown profound success in generating photo-realistic images. In contrast, standard HVAEs often produce images of inferior quality compared to…

HGEN: Heterogeneous Graph Ensemble Networks

arXiv:2509.09843v1 Announce Type: new Abstract: This paper presents HGEN that pioneers ensemble learning for heterogeneous graphs. We argue that the heterogeneity in node types, nodal features, and local neighborhood topology poses significant challenges for ensemble learning, particularly in accommodating diverse…

Input-Time Scaling

arXiv:2508.13654v4 Announce Type: replace Abstract: Current Large Language Models (LLMs) are usually post-trained on large-scale carefully curated datasets (data & training scaling) and doing reasoning in test time (inference time scaling). In this work, we present a new scaling paradigm,…

Meta-Learning Reinforcement Learning for Crypto-Return Prediction

arXiv:2509.09751v1 Announce Type: new Abstract: Predicting cryptocurrency returns is notoriously difficult: price movements are driven by a fast-shifting blend of on-chain activity, news flow, and social sentiment, while labeled training data are scarce and expensive. In this paper, we present…

LAVa: Layer-wise KV Cache Eviction with Dynamic Budget Allocation

arXiv:2509.09754v1 Announce Type: new Abstract: KV Cache is commonly used to accelerate LLM inference with long contexts, yet its high memory demand drives the need for cache compression. Existing compression methods, however, are largely heuristic and lack dynamic budget allocation.…