Menü
Skills in diesem Repository
SkillsMP hat 1.228 Skills aus ADu2021/skillXiv gesammelt. Öffne einen Skill, um Quelle und Details zu prüfen.
ADu2021/skillXivEs werden 40 von 1.228 gesammelten Skills angezeigt.
Implement Prefix Grouper to accelerate Group Relative Policy Optimization training by eliminating redundant prefix encoding, achieving up to 8x speedup for long-context scenarios.
Quellsprache: Englisch
Transform lengthy documents into fully narrated presentation videos with synchronized audio-visual delivery. Automatically segments content, generates visuals, synthesizes speech, and composes final video.
Quellsprache: Englisch
Bilevel 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.
Quellsprache: Englisch
Improve 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.
Quellsprache: Englisch
Understand 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…
Quellsprache: Englisch
Transform 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.
Quellsprache: Englisch
Phone 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…
Quellsprache: Englisch
Scale 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.
Quellsprache: Englisch
Unify 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…
Quellsprache: Englisch
Optimize 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…
Quellsprache: Englisch
Leverage 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.
Quellsprache: Englisch
Improves 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.
Quellsprache: Englisch
Learn 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.
Quellsprache: Englisch
Enable models to refine outputs dynamically during generation based on internal signals, reducing token consumption by 41.6% while improving accuracy by 8.2%.
Quellsprache: Englisch
Apply 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…
Quellsprache: Englisch
Improves 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…
Quellsprache: Englisch
Use 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.
Quellsprache: Englisch
Automatically 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…
Quellsprache: Englisch
Detect when diffusion language models converge on correct answers before completing refinement steps using confidence gap monitoring, achieving 3.4x decoding speedup
Quellsprache: Englisch
Train LLMs to discover novel reasoning strategies beyond base model capabilities using prolonged RL with KL control, reference policy resets, and diverse task suites.
Quellsprache: Englisch
Replaces 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.
Quellsprache: Englisch
Post-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…
Quellsprache: Englisch
Convert 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.
Quellsprache: Englisch
Enable 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.
Quellsprache: Englisch
Dramatically 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…
Quellsprache: Englisch
Combine 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.
Quellsprache: Englisch
Stabilize 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…
Quellsprache: Englisch
Fine-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…
Quellsprache: Englisch
Replace 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…
Quellsprache: Englisch
QueryBandits adaptively learns per-query rewriting strategies to reduce LLM hallucinations, achieving 87.5% improvement without model retraining.
Quellsprache: Englisch
Leverages 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.
Quellsprache: Englisch
State-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.
Quellsprache: Englisch
Systematically post-train models for long-context reasoning through multi-hop data synthesis, stabilized RL with adaptive entropy control, and memory-augmented architecture supporting 4M+ token sequences. Achieves performance comparable to GPT-5 and…
Quellsprache: Englisch
Construct multi-step reasoning benchmarks with interdependent problems to evaluate and improve long-horizon reasoning in large reasoning models. Enables evaluation of reasoning depth and breadth beyond single-step tasks.
Quellsprache: Englisch
Ground LLM world models with retrieved current knowledge from tutorials and documentation. Reduce hallucination in environment prediction and improve long-horizon planning by 16-23% on web agent benchmarks.
Quellsprache: Englisch
Co-evolutionary framework where Challenger generates tasks and Solver solves them. Models evolve autonomously from scratch without human annotations. Improves math reasoning +6.49pts and general reasoning +7.54pts.
Quellsprache: Englisch
Accelerate video diffusion models using sparse radial attention that exploits energy decay patterns. Achieves 3.7× speedup on long videos while maintaining quality through O(n log n) complexity instead of O(n²).
Quellsprache: Englisch
Optimize where multimodal models attend by treating attention weights as a learnable policy, using policy gradients with advantage weighting to improve visual grounding and perception without changing model architecture.
Quellsprache: Englisch
Learn optimal per-layer bit-width assignments for LLM quantization via RL, generalizing across models without retraining. Achieves superior compression under fixed bit budgets.
Quellsprache: Englisch
Apply rank-one weight modifications to amplify model safety via residual stream steering, requiring no fine-tuning and preserving utility on standard benchmarks
Quellsprache: Englisch