Tag
Explore by tags
Decouples perception and reasoning for hours-long videos by streaming inputs into a three-tier Hierarchical Graph Memory and using an agentic Observation–Reason–Action retrieval loop; reduces reasoning context to ~2% of full video while improving benchmark accuracy.
Simulates egocentric, embodied human–world interactions and enables customizable, self-evolving local scenes by defining anchor views and text-driven evolution. Uses exogenous viewpoints and full-body motion supervision to improve spatial grounding and interaction consistency.
A GGUF release of Gemma 4 26B A4B (QAT) packaged by Unsloth for local multimodal inference — quantization-aware trained to keep near-bfloat16 quality while significantly lowering memory requirements, compatible with Transformers and Unsloth tooling.
Models visual preference as distributions over rubric scores and introduces Z-Reward, a teacher–student framework that decouples reasoning-heavy judgment (teacher trained with GDSO) from efficient deployment (student via RISD). Demonstrates higher human-preference accuracy and works as a differentiable reward for text-to-image optimization.
A benchmark that evaluates interactive spatial reasoning for multimodal agents in realistic tasks. It unifies eight heterogeneous simulators under a simulator-agnostic protocol, provides 760 human-annotated tasks with vision-only partial observability, and uses text-based actions plus terminal-state verification to measure task success.
Stores a persistent 3D scene cache directly in a diffusion model's latent space to produce temporally and spatially consistent videos. Constructs memory via depth-guided back-projection and queries it with direct latent-space warping — achieving large speed and memory gains versus pixel-space 3D baselines.
Synthesizes scalable, photoreal 3D Earth tiles from georeferenced satellite imagery using a generative 3D Gaussian Splatting representation; trained on urban reconstructions, it generates novel scenes at under 10 minutes/km² with hierarchical LOD for real-time web map visualization and Embodied AI use cases.
End-to-end framework for controlled character animation that transfers motion from driving videos to reference characters without intermediate pose or background representations. Introduces the MotionPair‑60K end-to-end motion-transfer dataset, in‑context mask conditioning and mode‑specific RoPE for task unification, plus Bias‑Aware DPO to mitigate synthetic-detail errors.
Provides 500+ hours of human whole-body teleoperation demonstrations for humanoid robot learning in real homes, with synchronized video, joint states, action traces and language annotations. Includes 23K+ episodes, fine-grained subtask labels, and raw ROS/MCAP plus compressed LeRobot formats.
Generates text from interleaved text, image, and short-video inputs using discrete diffusion and block‑autoregressive multi‑canvas sampling; built on a sparse MoE (8/128) Gemma 4 backbone and optimized for low‑latency inference and very long contexts (up to 256K tokens).
Contains a sanitized Claude Code (Fable 5) JSONL transcript of a session that procedurally built a Boeing 747 in Three.js, including assistant messages, tool calls, and base64 screenshots — useful for studying agent trace, tool use, and vision self‑verification workflows.
Continuously watches live video and autonomously decides each second whether to speak, stay silent, or delegate; released together with an 8B vision-first model, time-aligned interaction data, training recipe, and a deployable real-time system. Designed for vision-triggered, low-latency streaming scenarios and evaluated across six real-world streams.