visual-control

Tag

Cards List
#visual-control

Directed Temporal Representations for Offline Visual Control

arXiv cs.LG ↗ · 2026-10-08 Cached

The paper introduces Directed Temporal Representations for Control (DTRC), which learns a directed temporal quasimetric over frozen LeWorldModel features from offline visual trajectories to serve as a goal-relative progress signal for direct goal-conditioned policy learning, achieving strong control performance across ten visual tasks.

0 favorites 0 likes
#visual-control

Reflex: Visually Grounded Reactive Humanoid Control

Hugging Face Daily Papers ↗ · 2026-10-08 Cached

Reflex提出一个三阶段框架,让类人机器人通过自我中心RGB-D视觉实现对抛掷物体的全身心反射式接取控制,在仿真中达到85.7%成功率,并在真实Unitree G1机器人上以2.5米距离实现65%的接取成功率。

0 favorites 0 likes
#visual-control

Anisotropic Representations Improve Planning in JEPA World Models

Hugging Face Daily Papers ↗ · 2026-09-29 Cached

This paper introduces AnisoWM, a method that improves planning in JEPA world models by using anisotropic representations with learnable covariance, leading to better task-aligned planning costs and enhanced performance in visual control environments.

0 favorites 0 likes
#visual-control

sensenova/SenseNova-U1.5-8B-MoT

Hugging Face Models Trending ↗ · 2026-08-19 Cached

SenseNova-U1.5-8B-MoT is a native unified multimodal model for enhanced visual creation, featuring improvements in image generation quality, text rendering, and precise control.

0 favorites 0 likes
#visual-control

Diagnosing JEPA World Models with Action-Conditioned Predictive Consistency

arXiv cs.LG ↗ · 2026-08-14 Cached

Introduces Action-Conditioned Predictive Consistency (ACPC), a diagnostic for JEPA world models that measures how clean and perturbed observations diverge under action-conditioned rollouts, with theoretical bounds on prediction error and planner cost. Experiments on visual control tasks validate the diagnostic across models like LeWM and PLDM.

0 favorites 0 likes
#visual-control

TaskSense: Focusing on What Matters in World Models

arXiv cs.AI ↗ · 2026-08-10 Cached

TaskSense introduces a task-centric world modeling framework that uses stochastic spatial attention conditioned on previous latent states and an auxiliary inverse-dynamics objective to focus on control-relevant regions, improving robustness to visual distractions compared to DreamerV3.

0 favorites 0 likes
#visual-control

Observation-Grounded Self-Predictive Reinforcement Learning for Visual Continuous Control

arXiv cs.LG ↗ · 2026-08-07 Cached

This paper proposes OG-SPR, a model-free visual RL algorithm that combines latent self-prediction with observation prediction to learn dynamics-aware representations, achieving improved sample efficiency on DeepMind Control Suite tasks.

0 favorites 0 likes
#visual-control

@CryptoTied: Guys! An AI coding agent emphasizing visual control has arrived. Juggler is an AI coding agent focused on "controllable, visual" interaction, with the core goal of giving developers true control over the LLM's interaction with the codebase, rather than black-box operation. It offers a native desktop app + Mi...

X AI KOLs Timeline ↗ · 2026-07-16 Cached

Juggler is an AI coding agent that emphasizes visualization and controllability. It provides a native desktop app, session tree structure, plugin architecture, and supports multiple models like Claude, OpenAI, Gemini, etc., allowing developers to clearly view and edit every step of the LLM's actions.

0 favorites 0 likes
← Back to home

Submit Feedback