Results for “supervised-fine-tuning”

Fine-tune instruction-following LLMs with Unsloth's optimized SFTTrainer, covering dataset formatting, chat templates, training configuration, and thinking-model patterns.

Fine-tune LLMs using LoRA, QLoRA, and PEFT with Hugging Face, including dataset preparation, hyperparameter tuning, evaluation, and deployment.

Fine-tune and align language models using reinforcement learning with TRL, including SFT, DPO, PPO, GRPO, and reward model training.

Visual Prompt Tuning

Trains and fine-tunes vision models for object detection, image classification, and segmentation using Hugging Face Transformers on cloud GPUs, with automatic dataset validation and Hub persistence.

Fine-tune and deploy CLIP vision-language models for zero-shot classification, image-text retrieval, and embedding extraction with ONNX and TensorRT support.

Fine-tune, evaluate, run inference, and export Cosmos-Embed1 video-text embedding models for tasks like text-to-video retrieval and semantic deduplication.

Fine-tune NV-Segment-CT VISTA3D on CT NIfTI labels for smoke testing or dataset adaptation, wrapping the upstream MONAI bundle entrypoint.

Diagnose and optimize slow-loading, high-memory, or low-FPS USD scenes using a structured workflow with profiling, validation, and mutation phases.

Finetunes the NV-Generate-CTMR MR-brain diffusion UNet from user-supplied NIfTI training volumes using a wrapper that stages configs and delegates to upstream scripts.

Finetune the NV-Generate-CTMR MAISI VAE/autoencoder on user-supplied CT or MRI NIfTI volumes using a staged config and datalist workflow.

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Automatically collects dialectic memory to fine-tune local models.

Computer vision engineering skill for object detection, image segmentation, and visual AI systems. Covers CNN and Vision Transformer architectures, YOLO/Faster R-CNN/DETR detection, Mask R-CNN/SAM segmentation, and production deployment with ONNX/TensorRT. Includes PyTorch, torchvision, Ultralytics, Detectron2, and MMDetection frameworks. Use when building detection pipelines, training custom models, optimizing inference, or deploying vision systems.

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Improved Baselines with Visual Instruction Tuning

Systematically repair a broken feature or module by scoping, tracing dependencies, diagnosing all issues, fixing them in order, and verifying end-to-end.

Scaling Instruction-Finetuned Language Models

Visual Instruction Tuning

Large Language Model development, training, fine-tuning, and deployment best practices.

Fine-tune large language models by training less than 1% of parameters using LoRA, QLoRA, and 25+ adapter methods, enabling efficient adaptation on consumer GPUs.

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

Refactor SwiftUI views toward small, explicit, stable view types with MV data flow, dedicated subviews, and correct Observation usage.

Instruction Tuning with GPT-4

Scaling Vision with Sparse Mixture of Experts

SVIT: Scaling up Visual Instruction Tuning

Mantis: Interleaved Multi-Image Instruction Tuning

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Guides analysis of Neuropixels recordings from raw data to curated units, covering preprocessing, motion correction, spike sorting, quality metrics, and export.

Provides expert guidance for distributed training with DeepSpeed, covering ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, and sparse attention.

>- Skill Refiner, the eval-guided improvement loop for SKILL.md files. Runs the bootstrap, score, propose, apply, and status cycle as a thin wrapper over the published @intentsolutions/refiner CLI, proposing safe, minimal, bounded SKILL.md edits and accepting an edit only when a held-out eval score strictly improves with no regression on any other case. Ships a 3-layer cost-tiered hook architecture (sinker, line, hook) that gates skill quality at edit time, end of turn, and commit time. Use when improving an existing skill, refining a SKILL.md against measured behavior, bootstrapping an eval set for a skill, or gating skill edits before they ship. Trigger with "/j-rig", "refine this skill", "bootstrap an eval set", "propose a skill edit", "promote the candidate", or "skill refiner status".

Guides cost-effective image generation by iterating at low resolution before producing final high-resolution output.

Diagnose and fix SwiftUI performance issues like slow rendering, janky scrolling, and high CPU/memory usage through code review and profiling guidance.