All authors
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab6,028 skills13 installs7,823 views
- Gedi Multiclass ControlBuild GeDi true/false control-code inputs for multi-class and zero-shot topic control with exactly two class-conditional passes.Votes: 0GitHub stars: 247
- Gedi PosteriorCompute GeDi generative-discriminator class posteriors from control and anti-control log probabilities using stable Bayes-rule normalization.Votes: 0GitHub stars: 247
- Gedi Recovery EvaluationRun and validate a bounded GeDi proxy recovery that invokes generated GeDi modules and records mechanism-faithful evidence.Votes: 0GitHub stars: 247
- Evaluation Harness RecoveryUse this skill when you need to compose task formatting, model adapter calls, filters, and metrics into an auditable evaluation run.Votes: 0GitHub stars: 247
- Metric AggregationUse this skill when you need to compute deterministic accuracy, exact match, token F1, and aggregate evaluation metrics.Votes: 0GitHub stars: 247
- Model Adapter InterfaceUse this skill when you need to wrap a model backend behind loglikelihood, rolling loglikelihood, and generation request methods.Votes: 0GitHub stars: 247
- Response FilteringUse this skill when you need to post-process raw model outputs with regex extraction, take-first, take-k, or majority-vote pipelines.Votes: 0GitHub stars: 247
- Task ConfigurationUse this skill when you need to define reproducible lm-evaluation-harness style task configurations and format documents into model requests.Votes: 0GitHub stars: 247
- Multitask Mixture SplitConstruct prompted multitask training mixtures while enforcing held-out dataset boundaries for zero-shot evaluation.Votes: 0GitHub stars: 247
- Prompt Template RenderingRender structured dataset examples into PromptSource-style natural-language source and target pairs for prompted multitask recovery.Votes: 0GitHub stars: 247
- Prompted Multitask TrainingExecute a deterministic minimal optimizer step on prompted multitask text examples and log loss/parameter changes.Votes: 0GitHub stars: 247
- Zero Shot Prompt Robust EvaluationCompute held-out prompted-task accuracy and prompt wording consistency for T0/P3-style recovery experiments.Votes: 0GitHub stars: 247
- Pplm Attribute ModelsUse when a recovery or implementation needs differentiable PPLM attribute scores for BoW or simple discriminator-style control.Votes: 0GitHub stars: 247
- Pplm Candidate RerankingUse when generated candidates must be ranked by attribute likelihood with Dist-n diversity filtering as in PPLM BCR.Votes: 0GitHub stars: 247
- Pplm Latent PerturbationUse when a PPLM-style controller must perturb latent/logit state at inference time with attribute and KL losses while freezing the base model.Votes: 0GitHub stars: 247
- Pplm Recovery EvaluationUse when validating a reduced or full PPLM recovery experiment with numeric metrics and mechanism checks.Votes: 0GitHub stars: 247
- Diffusion ScheduleBuild and validate DDPM forward-process schedules and closed-form noisy samples for reduced recovery experiments.Votes: 0GitHub stars: 247
- Epsilon ObjectiveCompute DDPM epsilon-prediction losses and weighted denoising objective terms for recovery training checks.Votes: 0GitHub stars: 247
- Reduced Recovery HarnessRun a bounded DDPM proxy experiment that exercises generated schedule, epsilon-loss, and reverse-step skills.Votes: 0GitHub stars: 247
- Reverse Denoising StepConvert predicted DDPM epsilon values into reverse-process means and deterministic or stochastic denoising samples.Votes: 0GitHub stars: 247
- Cfm Training ObjectiveCompute Conditional Flow Matching losses and bounded optimizer updates for a small parametric vector field.Votes: 0GitHub stars: 247
- Cnf Ode SamplingIntegrate learned Flow Matching vector fields with fixed-step ODE solvers and report trajectory/error diagnostics.Votes: 0GitHub stars: 247
- Conditional Flow Matching LossCompute Conditional Flow Matching squared vector-field losses and deterministic one-parameter proxy updates.Votes: 0GitHub stars: 247
- Conditional Ot PathsConstruct and validate Flow Matching conditional optimal-transport paths and target vector fields for paired noise/data samples.Votes: 0GitHub stars: 247
- Conditional Path BuilderBuild Gaussian and optimal-transport conditional Flow Matching path samples and target vector fields from noise/data pairs.Votes: 0GitHub stars: 247
- Flow Matching Recovery HarnessRun a bounded soft-mode Flow Matching proxy recovery that invokes generated OT-path, CFM-loss, and CNF-ODE skills.Votes: 0GitHub stars: 247
- Ode Sampler CheckerIntegrate simple Flow Matching vector fields and verify ODE trajectory, endpoint, and NFE evidence.Votes: 0GitHub stars: 247
- Proxy Recovery EvaluatorEvaluate whether a soft-mode reduced Flow Matching recovery is executable, metric-bearing, and mechanism-faithful.Votes: 0GitHub stars: 247
- I2sb Bridge PosteriorCompute the analytic Gaussian bridge posterior q(Xt | X0, X1) for I2SB paired clean/degraded endpoints.Votes: 0GitHub stars: 247
- Conditional Flow Matching ObjectiveBuild conditional flow matching interpolation states, target velocities, and squared regression losses for bounded CNF recovery experiments.Votes: 0GitHub stars: 247
- Minibatch Ot CouplingCompute deterministic squared-cost minibatch optimal transport pairings for OT conditional flow matching recovery runs.Votes: 0GitHub stars: 247
- Recovery Evaluation HarnessRun bounded executable OT-CFM proxy recoveries with command logs, mechanism checks, and validator-compatible artifacts.Votes: 0GitHub stars: 247
- Schrodinger Bridge CouplingBuild small entropy-regularized coupling checks and bridge variance schedules for SB conditional flow matching.Votes: 0GitHub stars: 247
- Msfm Batch CouplingConstruct uniform and exact small-batch BatchOT couplings for Multisample Flow Matching recovery and tests.Votes: 0GitHub stars: 247
- Msfm Joint Cfm LossCompute Joint Conditional Flow Matching interpolation, target velocity, loss, and tiny optimizer updates for MSFM recovery.Votes: 0GitHub stars: 247
- Msfm Mechanism EvaluationEvaluate whether an MSFM recovery result exercised BatchOT coupling, Joint CFM loss, optimizer evidence, and soft-mode proxy constraints.Votes: 0GitHub stars: 247
- Msfm Synthetic RecoveryRun a bounded synthetic MSFM proxy experiment that invokes generated coupling and Joint CFM loss skills.Votes: 0GitHub stars: 247
- Rectified Coupling InterpolationBuild rectified-flow coupling interpolation records with endpoint displacement targets for velocity regression.Votes: 0GitHub stars: 247
- Rectified Ode ReflowSimulate rectified-flow Euler transport and compute straightness, reflow, and convex transport diagnostics.Votes: 0GitHub stars: 247
- Rectified Recovery HarnessCompose rectified-flow module skills into an executable reduced recovery experiment with auditable JSON logs.Votes: 0GitHub stars: 247
- Rectified Velocity RegressionFit and audit a small rectified-flow velocity model with least-squares regression and optimizer evidence.Votes: 0GitHub stars: 247
- Continuous Score LossBuild continuous denoising score-matching targets and deterministic optimizer checks for score-SDE recovery.Votes: 0GitHub stars: 247
- Predictor Corrector SamplingExecute bounded reverse predictor-corrector sampling checks for score-SDE mechanism-faithful recovery.Votes: 0GitHub stars: 247
- Score Sde Recovery HarnessAssemble executable soft-mode recovery artifacts for reduced score-SDE training and mechanism validation.Votes: 0GitHub stars: 247
- Score Sde ScheduleDefine continuous-time score-SDE schedules, perturbation kernels, reverse drifts, and probability-flow checks for recovery experiments.Votes: 0GitHub stars: 247
- Conditional Pair ProtocolBuild and validate SR3-style low-resolution conditioning paired with high-resolution targets for reduced or full recovery experiments.Votes: 0GitHub stars: 247
- Diffusion Noise ObjectiveCompute SR3-style conditional diffusion corruption, noise-prediction loss, and a deterministic scalar optimizer update.Votes: 0GitHub stars: 247
- Iterative Refinement SamplerRun a deterministic SR3-inspired reverse denoising trajectory conditioned on low-resolution input for proxy recovery.Votes: 0GitHub stars: 247
- Proxy Recovery EvaluatorEvaluate whether an SR3 soft-mode proxy recovery exercised conditioning, diffusion loss, optimizer update, and iterative refinement evidence.Votes: 0GitHub stars: 247
- Sr3 Conditional Denoising LossCompute and optimize the SR3 conditional denoising objective for a lightweight trainable denoiser. This reusable skill supports bounded SR3 recovery experiments.Votes: 0GitHub stars: 247