Image, video, audio, and text generation
MIT Professor Rafael Gómez-Bombarelli discusses the transformative potential of AI in scientific research, emphasizing its role in materi...
![[2603.10202] Hybrid Hidden Markov Model for Modeling Equity Excess Growth Rate Dynamics: A Discrete-State Approach with Jump-Diffusion](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
Abstract page for arXiv paper 2603.10202: Hybrid Hidden Markov Model for Modeling Equity Excess Growth Rate Dynamics: A Discrete-State Ap...
Abstract page for arXiv paper 2602.00388: Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode
The paper presents SOM-VQ, a novel tokenization method that enhances interactive generative models by integrating vector quantization wit...
This article presents a novel approach to integrating deep unfolding techniques with MCMC methods, enhancing the efficiency and interpret...
The paper presents a novel estimator for model evidence in Bayesian inverse problems, particularly using diffusion models, enhancing samp...
The paper introduces Momentum Guidance (MG), a novel technique for enhancing flow-based generative models, achieving significant improvem...
This paper explores the dynamics of reasoning in Large Language Models (LLMs) through Manifold Capacity Theory, revealing how latent repr...
This paper presents a novel discrete denoising diffusion framework that utilizes a sample-efficient estimator for single-site conditional...
The paper presents KBVQ-MoE, a novel framework for improving vector quantization in Mixture of Experts (MoE) large language models, addre...
This paper introduces the Persona Brainstorm Audit (PBA), a method for assessing bias in Large Language Models (LLMs) used in creative ap...
This article presents MetamerGen, a novel tool that generates metamers of human scene understanding by combining low-resolution gist info...
This paper investigates safety alignment in large language models (LLMs) and large reasoning models (LRMs), identifying key factors that ...
The paper presents HiGR, a novel framework for generative slate recommendation that enhances efficiency and user preference alignment thr...
The paper introduces Refusal Steering, a method for controlling Large Language Models' refusal behavior on sensitive topics without retra...
This article presents a novel data-efficient approach for fine-tuning text-to-video generation models, demonstrating that low-quality syn...
The paper presents Latent-Augmented Discrete Diffusion Models (LADD), which enhance discrete diffusion models for improved language gener...
This article explores how rationales generated by large language models (LLMs) influence human judgments of plausibility in commonsense r...
The paper introduces Proportionate Credit Policy Optimization (PCPO), a novel framework aimed at improving the stability and quality of t...
The paper presents MCTD-ME, a novel approach combining Monte Carlo Tree Search and masked diffusion models for efficient protein design, ...
RooseBERT introduces a specialized language model for political discourse, enhancing the analysis of political debates through improved s...
DesignBench introduces a comprehensive benchmark for evaluating MLLM-based front-end code generation, addressing limitations in existing ...
HSSBench introduces a benchmark for evaluating Multimodal Large Language Models (MLLMs) in Humanities and Social Sciences, addressing gap...
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime