tao-generate-referring-expressions
Skill from NVIDIA/skills
Install
npx skills add NVIDIA/skills --skill tao-generate-referring-expressions
Runs the open skills CLI and installs the skill for your agent (Claude Code, Codex, Cursor …). Alternatively copy the skill folder into .claude/skills/ or upload it in Claude.ai under Settings → Skills.
What does the tao-generate-referring-expressions skill do?
Description from the SKILL.md in the source repository.
Questions about tao-generate-referring-expressions
What does the tao-generate-referring-expressions skill do?
Four-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region descriptions, scene captions, grounded referring expressions, and (optionally) verified expressions via VLM distillation. Use when the user wants to generate referring-expression annotations from images with KITTI labels, build region descriptions, produce grouped grounding phrases tied to bboxes, run a double-check verification pass on grounding expressions, auto-label traffic / scene images for referring datasets, or run the image_referring_expression pipeline. Triggers include 'referring expression', 'region description', 'KITTI labels', 'spatial relationship annotation', 'auto-label image referring expression', 'image_referring_expression'.
How do I install tao-generate-referring-expressions?
In a terminal: npx skills add NVIDIA/skills --skill tao-generate-referring-expressions. The skills CLI asks which agents to install it for. In Claude.ai you can upload the skill folder as a ZIP under Settings → Skills.
Which agents can use tao-generate-referring-expressions?
Skills use the open SKILL.md format, which Claude Code, Claude.ai, the Claude API, OpenAI Codex, Cursor, Gemini CLI and other agents support.
Is tao-generate-referring-expressions safe to use?
Our automatic check found no risky patterns (piped shell scripts, hidden characters, secret access, prompt overrides). It is not a full audit – read the SKILL.md and any scripts before installing.
More skills from NVIDIA/skills
Build and change NVIDIA ALCHEMI Toolkit (nvalchemi) workflows for atomistic simulation and machine-learned interatomic potentials (MLIPs).
Guide an agent through GPU-accelerating a codebase: setup interview, flavor-specific translation (Fortran OpenACC, Fortran OpenMP, C++ CUDA, C++ Kokkos), and NVIDIA Nsight Systems (nsys) profile-guided orchestration.
Set up, build, or debug physics simulations with the Newton Python package. Use when installing Newton, choosing simulation components, finding examples, or diagnosing simulation and runtime problems.
Measure or improve the runtime, startup time, memory use, or throughput of a Newton simulation that already behaves correctly, or review a Newton step loop for performance problems such as CUDA graph …
Validate NVIDIA FastGen integrations and training quality: numerical parity against reference implementations, seeded side-by-side media, gradient and trainability checks, checkpoint round trips and repository tests.
Run, resume or troubleshoot NVIDIA FastGen training: SFT or distillation with train.py, bounded smoke tests, distributed launches, checkpoints, validation and W&B evidence.
Set up or repair an NVIDIA FastGen environment for a chosen task: Docker or Conda installation, credentials and model access, weight and dataset downloads, and an optional bounded smoke test.
Resolve the execution context for an NVIDIA FastGen command: checkout and interpreter, config path and override syntax, output/cache/data roots, run identity, and distributed launch variables.
Related skills in Design & Media
Read first to make, edit, animate or render any video with HyperFrames (HTML rendered to video), including launch films, explainers, music videos, charts, maps, vertical clips, title cards, captions and overlays.
Get or make any media for a HyperFrames video - music, sound effects, images, icons, logos, voiceover, avatar and generated video, transcripts, cutouts, colour grades and looks - with HeyGen or free …
Design mobile app screens with Sleek, edit Sleek projects, and implement their designs in React Native or HTML.
A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
Generate AI videos with Google Veo, Seedance 2.0, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI.
Generate AI images with GPT-Image-2, FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI.
Create AI avatar and talking head videos via inference.sh CLI. Recommended: P-Video-Avatar (fastest, cheapest, built-in TTS). Also: OmniHuman, Fabric, PixVerse. Audio: Inworld TTS-2 (100+ languages, emotion steering for characters), ElevenLabs, Kokoro.
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable …