AI Infrastructure

UMKC Announces New Master of Science in Artificial Intelligence

UMKC announces a new Master of Science in Artificial Intelligence program aimed at addressing workforce demand for AI expertise, set to l...

Siemens, NVIDIA hit chip verification milestone for AI

Dell and HIVE partner to deploy Nvidia’s next-generation AI chips

[2602.16698] Causality is Key for Interpretability Claims to Generalise

This paper discusses the importance of causality in interpretability research for large language models, highlighting pitfalls in general...

[2602.15919] Generalized Leverage Score for Scalable Assessment of Privacy Vulnerability

The paper presents a method for assessing privacy vulnerability in machine learning models using a generalized leverage score, enabling e...

[2602.15902] Doc-to-LoRA: Learning to Instantly Internalize Contexts

The paper presents Doc-to-LoRA, a hypernetwork that enables Large Language Models to internalize contexts efficiently, reducing memory us...

[2602.16596] Sequential Membership Inference Attacks

The paper presents a novel approach to Membership Inference Attacks (MIAs) by developing an optimal attack strategy, SeMI*, leveraging mo...

[2602.15889] Evidence for Daily and Weekly Periodic Variability in GPT-4o Performance

This article investigates the temporal variability in the performance of the GPT-4o model, revealing significant daily and weekly pattern...

[2602.15888] NeuroSleep: Neuromorphic Event-Driven Single-Channel EEG Sleep Staging for Edge-Efficient Sensing

NeuroSleep presents a neuromorphic event-driven system for efficient EEG sleep staging, achieving high accuracy with reduced computationa...

[2602.16570] Steering diffusion models with quadratic rewards: a fine-grained analysis

This article presents a detailed analysis of sampling from reward-tilted diffusion models, focusing on quadratic rewards and their comput...

[2602.15862] Enhancing Action and Ingredient Modeling for Semantically Grounded Recipe Generation

This paper presents a novel framework for improving recipe generation from food images by enhancing action and ingredient modeling, addre...

[2602.15861] CAST: Achieving Stable LLM-based Text Analysis for Data Analytics

The paper presents CAST, a framework designed to improve the stability of LLM-based text analysis in data analytics by enhancing output c...

[2602.16507] Small molecule retrieval from tandem mass spectrometry: what are we optimizing for?

This paper explores the optimization of loss functions in deep learning models for small molecule retrieval from tandem mass spectrometry...

[2602.15858] State Design Matters: How Representations Shape Dynamic Reasoning in Large Language Models

This paper explores how state representations impact the reasoning capabilities of large language models (LLMs) in dynamic environments, ...

[2602.16498] Fast and Scalable Analytical Diffusion

The paper presents GoldDiff, a novel framework for analytical diffusion that enhances scalability and speed in generative modeling by dyn...

[2602.15856] Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector Perspective

The paper presents SeleCom, a novel selector-based soft compression framework for Retrieval-Augmented Generation (RAG), addressing limita...

[2602.16456] Beyond SGD, Without SVD: Proximal Subspace Iteration LoRA with Diagonal Fractional K-FAC

This paper presents a novel approach called LoRSum for optimizing Low-Rank Adaptation (LoRA) in machine learning, enhancing efficiency in...

[2602.15852] Building Safe and Deployable Clinical Natural Language Processing under Temporal Leakage Constraints

This article discusses the development of clinical NLP models that mitigate risks associated with temporal leakage, emphasizing the impor...

[2602.16442] Hardware-accelerated graph neural networks: an alternative approach for neuromorphic event-based audio classification and keyword spotting on SoC FPGA

The paper presents a hardware-accelerated graph neural network approach for neuromorphic event-based audio classification and keyword spo...

[2602.16400] Easy Data Unlearning Bench

The paper introduces the Easy Data Unlearning Bench, a unified benchmarking suite aimed at simplifying the evaluation of machine unlearni...

[2602.16363] Improved Bounds for Reward-Agnostic and Reward-Free Exploration

This paper presents improved algorithms for reward-free and reward-agnostic exploration in Markov decision processes, enhancing the abili...

[2602.15836] EdgeNav-QE: QLoRA Quantization and Dynamic Early Exit for LAM-based Navigation on Edge Devices

The paper presents EdgeNav-QE, a framework that combines QLoRA quantization and dynamic early exit mechanisms to enhance LAM-based naviga...

[2602.16340] The Implicit Bias of Adam and Muon on Smooth Homogeneous Neural Networks

This paper investigates the implicit bias of momentum-based optimizers like Adam and Muon in smooth homogeneous neural networks, extendin...