arXiv
UniSlider: Perceptually Uniform Sliders for Continuous Image Editing
Sliders provide an intuitive interface for continuous image editing. In current generative approaches, however, the s...
🤖 AI 前沿资讯
每天自动聚合 AI 领域最新动态
arXiv
Recursive Video In-Context Learning for Agentic Robot
LLM agents that orchestrate frozen vision-language-action (VLA) policies improve across episodes through text memory,...
arXiv
Learning to Read the Contextual Tokens in Diffusion Transformers
Multimodal Diffusion Transformers (MM-DiTs) jointly process visual and textual representations throughout generation....
arXiv
BiasFlow: Geometric Monitoring and Backbone Regularization for Spurious Feature Reliance
Worst-group accuracy (WGA) evaluates a trained predictor but does not characterize how its frozen backbone behaves wh...
arXiv
Base Models Can Reason By Taking a Cue From Training Data
In this paper, we study how training data creates associations between the tokens at the start of a base model's resp...
arXiv
One Figure, Every Canvas: Editable Flowchart Relayout via Agentic Pipeline
Pipeline figures in ML papers must be repurposed across many canvases, including paper columns, 16:9 slides, portrait...
HuggingFace
Code2Games: Enabling Coding Agents for Gaming World Generation
Generating a high-quality gaming world from a natural-language game intent requires joint reasoning about scene struc...
HuggingFace
QuantCode Model: Specializing Language Models for Executable Algorithmic Trading Code
Large language models are strong general-purpose code generators, but executable algorithmic trading remains a demand...
HuggingFace
From Knowledge Access to Source Learning: Developing Source-Specific Competence
Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-in...
HuggingFace
TextReg: Mitigating Prompt Distributional Overfitting via Regularized Text-Space Optimization
Large language models (LLMs) are highly sensitive to the prompts used to specify task objectives and behavioral const...
HuggingFace
PerturBot: Breaking Shortcut Priors in Vision-Language-Action Models with Perturbative Training
A vision--language--action (VLA) policy can complete complex tasks while ignoring the evidence that should determine ...
HuggingFace
Representation-Space MMD for Diffusion Language Models
We introduce a post-training method for diffusion language models (DLMs) that minimizes Maximum Mean Discrepancy (MMD...