القائمة
Skills في هذا المستودع
جمع SkillsMP عدد ١٬٢٢٨ من skills من ADu2021/skillXiv. افتح أي skill لمراجعة مصدره وتفاصيله.
ADu2021/skillXivعرض ٤٠ من أصل ١٬٢٢٨ skills مجمعة.
Combines DeepSeek Sparse Attention (DSA) achieving O(Lk) complexity, scalable RL framework allocating 10%+ compute to post-training, and large-scale agentic task synthesis with 1,800+ environments. Matches proprietary models in reasoning;…
لغة النص الأصلي: الإنجليزية
Enable efficient speculative decoding by training discrete diffusion language models for parallel draft generation. Use AR-style distillation and scribe refinement to train dLLMs. Eliminate left-to-right error accumulation through independent parallel…
لغة النص الأصلي: الإنجليزية
Improve diffusion model alignment by assigning step-wise rewards during denoising instead of terminal rewards. Fixes sparse reward signal mismatch in multi-step generation processes through ODE-based reward estimation.
لغة النص الأصلي: الإنجليزية
Extend monocular depth models to handle adverse conditions (weather, darkness, sensor noise) using only 540K training samples. Applies perturbation-based consistency learning and spatial constraints to maintain robust depth prediction across challenging…
لغة النص الأصلي: الإنجليزية
Replace LayerNorm with Derf(x) = erf(αx + s) for improved generalization in transformers. Derf outperforms LayerNorm across vision, speech, and DNA modeling—ideal when normalization-free training provides benefits without architectural complexity.
لغة النص الأصلي: الإنجليزية
Accelerate LLM inference 6x by using block diffusion for parallel token drafting with tight coupling to the target model's hidden representations, achieving higher speedups than existing speculative methods without quality loss.
لغة النص الأصلي: الإنجليزية
Unlocks parallelism in recurrent memory transformers through diagonal batching of the layers-segments grid, achieving 3.3x speedup on 131K-token sequences without model retraining.
لغة النص الأصلي: الإنجليزية
Train diffusion language models to generate optimized CUDA kernels using bi-phase reinforcement learning. First phase masks and regenerates core kernel logic with provided scaffolding to prevent PyTorch shortcuts. Second phase enables end-to-end generation.…
لغة النص الأصلي: الإنجليزية
Recast chain-of-thought reasoning as iterative denoising using diffusion principles to overcome exposure bias in autoregressive reasoning. DiffCoT enables retrospective refinement of intermediate steps while maintaining temporal consistency through causal…
لغة النص الأصلي: الإنجليزية
Improve state-space model performance on language modeling and retrieval by applying differential mechanisms to reduce noise in intermediate representations and enhance long-context reasoning.
لغة النص الأصلي: الإنجليزية
Apply diffusion models as native generative agents for vision-centric reasoning tasks (sequential planning, constraint satisfaction, spatial configuration) instead of text-based LLM chains. Achieves 3x+ improvements over GPT-5 and Gemini-3 on visual…
لغة النص الأصلي: الإنجليزية
Train masked diffusion models for code generation using coupled-GRPO to optimize non-autoregressive generation. Achieves 4.4% improvement on code benchmarks while reducing autoregressive bias.
لغة النص الأصلي: الإنجليزية
Extract fine-grained visual features from pretrained text-to-image diffusion models to improve multimodal LLMs beyond CLIP, achieving up to 6% gains through strategic fusion of unconditional and conditional diffusion representations.
لغة النص الأصلي: الإنجليزية
Convert pre-trained autoregressive vision-language models into diffusion VLMs without architectural modifications. Use block diffusion strategy enabling arbitrary-length generation and KV-cache reuse. Hybrid attention enforces bidirectional within blocks,…
لغة النص الأصلي: الإنجليزية
Research contribution advancing agent and reasoning capabilities through novel approaches to model development, training, and evaluation.
لغة النص الأصلي: الإنجليزية
Apply reasoning-focused RL to vision foundation models using GRPO to develop deep visual understanding and abstract reasoning beyond visual recognition.
لغة النص الأصلي: الإنجليزية
Improve diffusion model capacity efficiency by directly predicting clean data instead of noise—leverage the manifold assumption that natural data occupies low-dimensional space while noise spans full dimensionality.
لغة النص الأصلي: الإنجليزية
Optimize diffusion models for preference alignment by learning directly from group-level preferences without stochastic policies, enabling efficient and stable training.
لغة النص الأصلي: الإنجليزية
Enable effective RL for diffusion language models via DiPO (unbiased GRPO for dLLMs) and framework optimizations. FlexAttention accelerates blockwise training, LMDeploy optimizes inference, achieving training-inference consistency—improving dLLM math…
لغة النص الأصلي: الإنجليزية
Enables diffusion LLMs to achieve 2.5× faster inference than autoregressive models through block-wise generation with parallel inter-block decoding.
لغة النص الأصلي: الإنجليزية
Reformulate discrete diffusion using signal-to-noise ratio for hybrid masked-uniform noise scheduling. Derive compute-optimal scaling laws through careful hyperparameter tuning, showing uniform diffusion scales more favorably in token-constrained settings…
لغة النص الأصلي: الإنجليزية
Understand fundamental limitations of applying diffusion to discrete text: position-agnostic corruption ignores linguistic structure, and token-wise training misses multi-token dependencies. Design text diffusion systems satisfying five essential properties:…
لغة النص الأصلي: الإنجليزية
Evaluate and optimize tool-integrated reasoning in LLMs through empirical benchmarking, performance-cost metrics (PAC, AUC-PCC), and measurement frameworks for diverse reasoning tasks.
لغة النص الأصلي: الإنجليزية
Convert inference-time feedback into persistent, retrievable guidelines stored as agent memory. Framework enables LLMs to improve performance over time by systematically accumulating and applying learned critiques. Augmented models rapidly match test-time…
لغة النص الأصلي: الإنجليزية
Align latent distributions with arbitrary reference distributions via explicit matching constraints rather than fixed priors. DMVAE achieves gFID 3.2 on ImageNet with 64 epochs—when you need flexibility in latent representation design for image generation.
لغة النص الأصلي: الإنجليزية
Synthesize diverse, verifiable training tasks by executing real tools first, then reverse-deriving tasks from execution traces. Ensure diversity across tools and reasoning patterns while maintaining grounding by construction.
لغة النص الأصلي: الإنجليزية
Train models to decompose complex problems into subproblems via divide-and-conquer reasoning. Structured approach enables systematic solution assembly and improved long-horizon reasoning compared to end-to-end generation.
لغة النص الأصلي: الإنجليزية
Improve multimodal reasoning at test-time through confidence-guided latent optimization without retraining. Iteratively refine learnable latent think tokens via policy gradient using confidence reward. Dynamically select and update relevant image patches…
لغة النص الأصلي: الإنجليزية
Generate multiple tokens simultaneously by having late transformer layers directly predict multiple outputs after early layer processing. No auxiliary parameters or verification needed. Achieve up to 2x generation speedup.
لغة النص الأصلي: الإنجليزية
Systematic evaluation toolkit for assessing large language models across multiple dimensions, enabling comprehensive benchmarking of agent capabilities and comparative analysis of model performance.
لغة النص الأصلي: الإنجليزية
Build open-source agents for document question-answering by modeling DocQA as information-seeking with explicit tool utilization. DocDancer uses an exploration-then-synthesis pipeline to generate high-quality training data, addressing the scarcity that limits…
لغة النص الأصلي: الإنجليزية
Improve long-context length extrapolation by denoising instabilities in Rotary Position Embeddings (RoPE) through spectral analysis and selective head rewriting—training-free post-hoc intervention for longer context windows.
لغة النص الأصلي: الإنجليزية
Compress LLMs by 20-30% in width while preserving functionality through optimal transport-based neuron merging. Instead of discarding neurons, redistribute their signal to retained neurons via learned transport maps. Use when you need to reduce model size…
لغة النص الأصلي: الإنجليزية
Diagnose and fix multi-agent system failures through targeted interventions (message edits, plan changes) rather than static log analysis. DoVer recovers 18-28% of failed trials with 30-60% hypothesis validation—essential for autonomous multi-agent…
لغة النص الأصلي: الإنجليزية
Enable Diffusion Language Models to achieve 8.5x inference speedup (24-30 steps vs. 256) through certainty-forcing distillation that trains models to achieve simultaneous high confidence across multiple tokens. Use when optimizing inference latency for dLLM…
لغة النص الأصلي: الإنجليزية
Replace PPO's heuristic ratio-based clipping with Divergence Proximal Policy Optimization (DPPO) that directly constrains policy divergence using either Total Variation or KL, enabling lightweight approximations (Binary, Top-K) for vocabulary-scale…
لغة النص الأصلي: الإنجليزية
Implements DPWriter from arXiv:2601.09609
لغة النص الأصلي: الإنجليزية
Use per-layer routers trained with Monte Carlo Tree Search to dynamically skip, execute, or repeat transformer layers for each token. Saves ~5 layers per example while maintaining accuracy on diverse benchmarks.
لغة النص الأصلي: الإنجليزية
Enable stable multi-agent reinforcement learning by normalizing advantages per-agent rather than globally, preventing gradient-norm inflation in heterogeneous multi-agent systems.
لغة النص الأصلي: الإنجليزية
Dr. Zero: Self-Evolving Search Agents without Training Data. From arXiv:2601.07055
لغة النص الأصلي: الإنجليزية