All authors
Claude Skills by ADu2021
github.com/ADu20211,227 skills5 installs2,395 views
- Perk Long Context Test Time LearningEnable language models to reason over extremely long contexts (128K tokens) by encoding context into lightweight LoRA adapters during test time, achieving 20% performance improvements without full model retraining.Votes: 0GitHub stars: 6
- Persona Vectors Llm Monitoring ControlMethod to identify and manipulate interpretable directions in model activation space corresponding to personality traits. Enables real-time monitoring of trait drift and preventive steering to mitigate unwanted behavioral shifts in language models.Votes: 0GitHub stars: 6
- Pets Trajectory Allocation FrameworkOptimize sampling budget allocation for self-consistency inference by treating trajectory allocation as a crowdsourcing problem. Introduce self-consistency rate (agreement with infinite-budget consensus) as optimization target. Offline setting uses Optimistic Knowledge Gradient for Bayesian optimization; online setting uses difficulty grid discretization and greedy allocation. Achieves up to 75% budget reduction (offline) and 55% (online) while maintaining accuracy.Votes: 0GitHub stars: 6
- Physrvg Physics Aware Unified ReinforcementPhysical principles are fundamental to realistic visual simulation, but remain a significant oversight in transformer-based video generation. This gap highlights a critical limitation in rendering rigid body motion, a core tenet of classical mechanics. While computer graphics and physics-based simulators can easily model such collisions using Newton formulas, modern pretrain-finetune paradigms discard the concept of object rigidity during pixel-level global denoising. Even perfectly correct m...Votes: 0GitHub stars: 6
- Pi Rl Flow Matching Vla Fine TuningApply reinforcement learning to flow-based VLA models by modeling denoising as an MDP and converting ODEs to SDEs, enabling efficient policy optimization for robotic control without expensive supervised fine-tuning data collection.Votes: 0GitHub stars: 6
- Pisa Sparse AttentionAccelerate diffusion transformers through training-free sparse attention combining exact computation for critical blocks with efficient approximation for non-critical ones. Achieves 2-3× speedup without retraining.Votes: 0GitHub stars: 6
- Pixio Masked AutoencoderEnhance Masked Autoencoders through three algorithmic improvements: deeper decoders enabling semantic-focused encoding, larger masking blocks providing richer context, and multiple class tokens capturing diverse global properties. Combine with 2B web-crawled images and soft self-curation for competitive downstream performance.Votes: 0GitHub stars: 6
- Plan Budget ReasoningDecompose complex queries into sub-questions and allocate computational budgets adaptively based on estimated difficulty, achieving 70% accuracy improvements and 39% token reduction without retraining.Votes: 0GitHub stars: 6
- Play To GeneralizeDevelop reasoning capabilities in multimodal models through interactive gameplay rather than supervised learning, achieving superior transfer to diverse reasoning tasks.Votes: 0GitHub stars: 6
- Plenoptic Video GenerationGenerate spatially and temporally coherent multi-view video through autoregressive conditioning with camera-guided retrieval and progressive context scaling. Enable long-video generation maintaining spatio-temporal memory across viewpoint changes.Votes: 0GitHub stars: 6
- Poet Orthogonal Llm TrainingImprove LLM training stability and generalization by reparameterizing weight matrices as orthogonal transformations, achieving better perplexity than AdamW with fewer trainable parameters.Votes: 0GitHub stars: 6
- Polca Stochastic Generative OptimizationUse language models as optimizers to iteratively improve complex systems (prompts, code, agent policies) through noisy feedback and sampling. POLCA maintains a priority queue of candidates, explores with epsilon-nets, and learns meta-insights from trial history.Votes: 0GitHub stars: 6
- Policy Discriminative Reward ModelsLearn generalizable reward models via unsupervised policy discrimination: pretraining models to distinguish between different policies enables efficient adaptation to human preferences and strong RLHF performance.Votes: 0GitHub stars: 6
- Pop Prefill Only PruningRemove deep layers during context encoding (prefill) while keeping them for token generation (decode). Identifies layer importance asymmetry via virtual gates; achieves 1.37x prefill speedup without retraining on any pre-trained model.Votes: 0GitHub stars: 6
- Poss Speculative DecodingImprove speculative decoding throughput by employing position-specialized draft layers that handle position-specific error accumulation patterns.Votes: 0GitHub stars: 6
- Practitioner Guide Multi Turn Agentic RlTrain LLM agents via multi-turn reinforcement learning by systematically optimizing environment complexity, reward signals, and policy initialization. Use curriculum learning, dense verified rewards, and domain-specific SFT for reliable agent convergence across TextWorld, ALFWorld, and SWE-Gym benchmarks.Votes: 0GitHub stars: 6
- Pref Grpo Text To ImageOptimize text-to-image models using pairwise preference comparisons instead of pointwise rewards, eliminating normalization instability and reward hacking while providing fine-grained image quality differentiationVotes: 0GitHub stars: 6
- Prefix Grouper GrpoImplement Prefix Grouper to accelerate Group Relative Policy Optimization training by eliminating redundant prefix encoding, achieving up to 8x speedup for long-context scenarios.Votes: 0GitHub stars: 6
- Presentation Video GenerationTransform lengthy documents into fully narrated presentation videos with synchronized audio-visual delivery. Automatically segments content, generates visuals, synthesizes speech, and composes final video.Votes: 0GitHub stars: 6
- Pretrain Zero Active PretrainingBilevel min-max optimization where mask generator selects informative spans from pretraining data and mask predictor recovers them via chain-of-thought, enabling effective RL pretraining on noisy corpora without supervised fine-tuning.Votes: 0GitHub stars: 6
- Pretraining Data RefinementImprove pretraining efficiency by refining noisy data through expert-guided programs: learn to generate deletion operations that clean documents, achieving 2.6-7.2% performance gains with fewer training tokens.Votes: 0GitHub stars: 6
- Pretraining Midtraining Rl InterplayUnderstand when RL genuinely expands reasoning beyond pre-training through controlled experiments on synthetic tasks. Discover that RL works best at the edge of competence and process rewards reduce hacking—critical for designing effective reasoning model training.Votes: 0GitHub stars: 6
- Primo R1 Vision CriticTransform video multimodal models into active process critics for robotic tasks. Use RL to incentivize explicit reasoning about progress toward goals and anchor reasoning temporally between initial and current states.Votes: 0GitHub stars: 6
- Prism Benchmarking Phone Realization In SpeechPhone recognition (PR) serves as the atomic interface for language-agnostic modeling for cross-lingual speech processing and phonetic analysis. Despite prolonged efforts in developing PR systems, current evaluations only measure surface-level transcription accuracy. We introduce PRiSM, the first open-source benchmark designed to expose blind spots in phonetic perception through intrinsic and extrinsic evaluation of PR systems. PRiSM standardizes transcription-based evaluation and assesses dow...Votes: 0GitHub stars: 6
- Prism Diffusion ScalingScale inference efficiency for discrete diffusion language models through hierarchical trajectory search with adaptive pruning and self-verified feedback. Achieve 3-4× speedup versus best-of-N with equal quality.Votes: 0GitHub stars: 6
- Prism HypothesisUnify semantic understanding and pixel-level detail in a single representation by decomposing features into frequency bands. Low frequencies encode semantics while high frequencies capture pixels—enabling one tokenizer for both understanding and generation through frequency-based modulation and semantic-wise alignment.Votes: 0GitHub stars: 6
- Prism Process Reward Guided InferenceOptimize multi-step reasoning by treating candidate solutions as particles in a process-reward energy landscape. Use PRM step-level scores to guide stochastic refinement and population resampling, achieving directional error correction without hallucination amplification.Votes: 0GitHub stars: 6
- Privileged Information Object DetectionLeverage training-time privileged information (depth, saliency maps) to improve student detector performance without inference overhead. Model-agnostic methodology applicable across detection architectures with no increase in inference complexity.Votes: 0GitHub stars: 6
- Prl Process Reward LearningImproves LLM reasoning by decomposing RL objectives into intermediate process rewards assigned to reasoning steps, improving both final accuracy and reasoning capacity without expensive Monte Carlo Tree Search.Votes: 0GitHub stars: 6
- Proact Agentic LookaheadLearn to ground LLM agent planning in real environment dynamics using Monte-Carlo Tree Search exploration combined with lightweight Monte-Carlo critics, reducing hallucination-driven planning failures in interactive tasks.Votes: 0GitHub stars: 6
- Proactive Self RefinementEnable models to refine outputs dynamically during generation based on internal signals, reducing token consumption by 41.6% while improving accuracy by 8.2%.Votes: 0GitHub stars: 6
- Profuse 3d Semantic UnderstandingApply semantic understanding to 3D Gaussian Splatting scenes through dense correspondence-guided pre-registration without render-supervised fine-tuning. Achieve semantic understanding in ~5 minutes using cross-view clustering and direct language feature fusion.Votes: 0GitHub stars: 6
- Progressive Residual WarmupImproves LLM convergence and downstream task performance by introducing time-dependent scaling to residual connections, enabling shallow layers to learn first before deeper layers activate. Apply during model pretraining to achieve 0.4-4.86 perplexity reduction.Votes: 0GitHub stars: 6
- Prompt Orchestration MarkupUse component-based markup with CSS-like styling to structure complex prompts, integrate diverse data types, and separate content from formatting for maintainable, version-control-friendly LLM applications.Votes: 0GitHub stars: 6
- Promptquine Evolving In Context PromptsAutomatically discovers optimal in-context learning prompts through evolutionary token pruning that removes redundant demonstrations to create effective 'gibberish' prompts. Matches state-of-the-art optimization with low-data regimes. Use for automated prompt discovery without manual tuning or human expertise.Votes: 0GitHub stars: 6
- Prophet Diffusion LmDetect when diffusion language models converge on correct answers before completing refinement steps using confidence gap monitoring, achieving 3.4x decoding speedupVotes: 0GitHub stars: 6
- Prorl Reasoning ExpansionTrain LLMs to discover novel reasoning strategies beyond base model capabilities using prolonged RL with KL control, reference policy resets, and diverse task suites.Votes: 0GitHub stars: 6
- Psa Pyramid Sparse AttentionReplaces binary keep/drop masks with multi-level pooled key-value representations, allowing queries to access larger receptive fields under same compute budget through hierarchical aggregation without discarding information.Votes: 0GitHub stars: 6
- Puzzle Curriculum GrpoPost-train vision-language models using automatically-verifiable puzzle environments (Jigsaw, Rotation, PatchFit) with graded rewards. Implement exploration-aware curriculum combining difficulty weighting with solution-space diversity metrics. Track reasoning-answer consistency to prevent divergence during training.Votes: 0GitHub stars: 6
- Pvchat Personalized Video One ShotEnable identity-aware video question answering with one-shot learning using Mixture-of-Heads enhanced ViLLM. Learns subject-specific features from single video through synthetic augmentation and progressive image-to-video training, enabling recognition of individuals in medical, smart home, and entertainment contexts.Votes: 0GitHub stars: 6
- Pyramidal Wan Video EfficiencyConvert pretrained video diffusion models into pyramidal architectures via low-cost finetuning while preserving output quality. Explore step distillation for enhanced efficiency, enabling deployment of efficient inference without training from scratch.Votes: 0GitHub stars: 6
- Pyvision Agentic Vision Dynamic ToolsEnable multimodal language models to autonomously generate and execute Python-based tools during visual reasoning, boosting performance on vision benchmarks by up to 31% through interactive problem-solving without relying on predefined tool sets.Votes: 0GitHub stars: 6
- Q Tuning Joint Pruning Efficient TrainingDramatically reduce training data requirements (to 12.5% of original) while improving model performance using joint sample and token pruning guided by Error-Uncertainty plane diagnostics. Asymmetric pruning preserves calibration signals while removing redundant tokens from misconception examples.Votes: 0GitHub stars: 6
- Qerl QuantizationCombine NFVP4 quantization with LoRA to accelerate RL rollout phases while using quantization noise as implicit exploration bonus. Achieve 1.5x speedup and better strategy discovery through noise-enhanced policy entropy.Votes: 0GitHub stars: 6
- Quantile Advantage Entropy Safe ReasoningStabilize LLM reasoning training by replacing mean-based advantage baselines with K-quantile baselines, preventing both entropy collapse and explosion while improving performance on mathematical benchmarks through response-level gating and asymmetric sample weighting.Votes: 0GitHub stars: 6
- Quantized Evolution StrategiesFine-tune quantized LLMs directly in low-precision discrete parameter space using evolution strategies with accumulated error feedback. Overcome gradient stagnation in quantized models by accumulating fractional updates using Delta-Sigma modulation, achieving significant improvements in INT4 quantized models without full-precision gradients.Votes: 0GitHub stars: 6
- Quco RagReplace unreliable model-internal confidence signals with objective corpus statistics to decide when RAG retrieval is necessary. Pre-evaluates entity rarity in training data and verifies entity co-occurrence at runtime, triggering retrieval only when hallucination risk is high—improving reliability without per-model tuning.Votes: 0GitHub stars: 6
- Query Bandits HallucinationQueryBandits adaptively learns per-query rewriting strategies to reduce LLM hallucinations, achieving 87.5% improvement without model retraining.Votes: 0GitHub stars: 6
- Qwen3 EmbeddingLeverages Qwen3 foundation models for text embedding and reranking via multi-stage training combining weakly-supervised pre-training on 150M synthetic pairs with supervised fine-tuning.Votes: 0GitHub stars: 6
- Qwen3 Vl MultimodalState-of-the-art multimodal model advancing vision-language understanding and generation capabilities through improved visual encoders, dense token representations, and unified reasoning over images and text.Votes: 0GitHub stars: 6