Menu
Skills dans ce dépôt
SkillsMP a collecté 1 228 skills depuis ADu2021/skillXiv. Ouvrez un skill pour examiner sa source et ses détails.
ADu2021/skillXivAffichage de 40 skills collectés sur 1 228.
To teach robots complex manipulation tasks, it is now a common practice to fine-tune a pre-trained vision-language-action model (VLA) on task-specific data. However, since this recipe updates existing representations, it is unsuitable for long-term operation…
Langue du texte source : anglais
Accelerate LLM inference by dynamically skipping transformer layers based on in-context signals, without training auxiliary draft models or changing model weights.
Langue du texte source : anglais
Represent 3D scenes as compressed light-field tokens for efficient neural rendering. Multi-view images are tokenized via Plücker coordinates, condensed through K-means clustering, and rendered adaptively. Achieves 5-7× data reduction versus MVSplat while…
Langue du texte source : anglais
Augment verifiable reward RL (RLVR) with contrastive learning to generate dense auxiliary rewards. Enforce proximity among correct reasoning trajectories in embedding space while suppressing errors, amplifying invariant reasoning patterns.
Langue du texte source : anglais
Replace hard clipping in policy gradients with smooth quadratic penalties derived from Total Variation divergence constraints. Eliminates zero-gradient regions and training instability while maintaining stable policy evolution.
Langue du texte source : anglais
No More Stale Feedback: Co-Evolving Critics for Open-World Agent Learning. From arXiv:2601.06794
Langue du texte source : anglais
Hybrid multi-agent architecture where orchestrator delegates tasks to GUI Operator or Programmer agent. Coding enables efficiency on computational tasks, achieving 60.76% on OSWorld with 33% fewer steps.
Langue du texte source : anglais
Dynamically allocate training budget across samples using a capability-oriented value function that measures per-sample training importance based on model capability evolution. Reduces training time via greedy heap-based allocation optimizing…
Langue du texte source : anglais
Generates complex text-heavy and structured images by converting natural language descriptions into executable code that specifies layouts, then renders and refines. Achieves 68% improvement on structured image generation benchmarks.
Langue du texte source : anglais
Train dual-brain agents with specialized planner (Cerebrum) and executor (Cerebellum) through decoupled RL, resolving planning-execution trade-off for scientific GUI agents
Langue du texte source : anglais
Train code and test generators through adversarial co-evolution where test LLM generates adversarial test cases to expose code defects. Prevent self-collusion by separating models and enabling white-box test generation.
Langue du texte source : anglais
Predict GUI state evolution by generating HTML code rather than pixel images. Combines visual fidelity of pixel-based approaches with structural precision of code-based methods through deterministic rendering. Enables agents to evaluate action consequences…
Langue du texte source : anglais
Generate physically grounded 4D scenes from natural language through dual-stream architecture separating object detail from scene orchestration. VLM-Motion Critic validates physics parameters iteratively, bridging semantic-physical execution gap.
Langue du texte source : anglais
Generate Verilog hardware code from natural language using reasoning-enhanced LLMs, combining rule-based testbench generation with round-trip data synthesis and adaptive DAPO reinforcement learning for reliable hardware design.
Langue du texte source : anglais
Evaluate LLM coding capabilities across three dimensions—problem analysis, code implementation, and test validation—to identify specific reasoning gaps and improve model robustness through ensemble approaches.
Langue du texte source : anglais
Advanced reasoning approach for optimizing inference efficiency through meta-cognitive planning, enabling agents to make better decisions with reduced computational overhead.
Langue du texte source : anglais
Align VLA efficiency with human cognition through 3-stage progressive routing: instruction-aware aggregation, instruction-irrelevant pruning, and coupled attention for 2.8x inference speedup
Langue du texte source : anglais
Dynamically adapt LLM depth per input at test time by skipping, repeating, or reordering layers using MCTS search, correcting 60% of initially wrong predictions and processing 75% of correct predictions with shorter architectures without retraining.
Langue du texte source : anglais
Iteratively retrieve and reason over long narratives using a dynamic memory workspace that integrates retrieved facts into a shared context for complex multi-hop reasoning.
Langue du texte source : anglais
Organize agent memory as an event graph with explicit logical relationships rather than flat embeddings. Framework incrementally segments experiences into events and links them through causal, temporal, and logical relations. Enables agents to navigate memory…
Langue du texte source : anglais
Build generalist judge models for evaluating LLM outputs using verifiable rewards and policy gradient training. Create a 7B model competitive with much larger judges through reward-guided optimization and critical thinking decomposition. Use when you need…
Langue du texte source : anglais
Lightweight verifier model for evaluating LLM outputs across multiple domains, handling diverse answer types through meta-error pattern analysis.
Langue du texte source : anglais
Improve RL sample efficiency through co-evolution of policy and experience extractor, enabling dynamic experience replay that adapts to the agent's skill level.
Langue du texte source : anglais
Decompose complex image editing instructions into simpler sub-tasks with automatically generated control guidance. Handles multi-object edits, preserves identity of surrounding regions, and eliminates manual mask creation.
Langue du texte source : anglais
Convert easy, high-accuracy training prompts into harder compositional problems by sequentially chaining multiple prompts together. Use Composition-RL to maintain effective learning signals during RL training when many prompts achieve near-perfect accuracy,…
Langue du texte source : anglais
Enable computer-use agents to flexibly choose between GUI primitives (click, type) and high-level tool calls, reducing cascading errors by 22% and improving execution speed by 11%.
Langue du texte source : anglais
Enable AI agents to safely explore action outcomes before execution by predicting UI state changes in desktop applications. Two-stage approach: first predict textual description of what changes, then synthesize visual representation of resulting screen.…
Langue du texte source : anglais
Learns richer spatial representations by training on both 2D and 3D data simultaneously. Combines 3D intra-modal self-distillation with 2D-3D cross-modal joint embedding, achieving 14.2% and 4.8% improvements over single-modality baselines in scene…
Langue du texte source : anglais
Improve language model reasoning using only model confidence as reward signals, eliminating need for labels or preference models while achieving substantial gains with minimal data.
Langue du texte source : anglais
Learn optimal configurations for agentic AI systems through hierarchical RL that treats configuration as a query-wise decision problem. Structure policy selects workflows/tools/budgets while prompt policy composes specific instructions, achieving 25% accuracy…
Langue du texte source : anglais
Improve student model robustness under covariate shift by using diffusion-based augmentation that targets spurious features via teacher-student disagreement.
Langue du texte source : anglais
Improves speculative decoding acceptance rates by exposing target model's intermediate reasoning through contemplate tokens. Achieves 8-11% acceptance rate improvement over EAGLE through future-direction guidance without extra forward passes.
Langue du texte source : anglais
Build AI code agents that scale to massive repositories with long-context reasoning and persistent memory. Confucius SDK achieves 59% Resolve@1 on SWE-Bench-Pro—ideal when AI needs to handle real-world codebases with complex toolchains.
Langue du texte source : anglais
Train adaptive ODE solvers that generate fast diffusion previews while maintaining consistency with full-step refinement. Learn context-aware integration coefficients through PPO without distilling base models. Achieve high-quality previews in few steps with…
Langue du texte source : anglais
Train a specialized continue-thinking token via reinforcement learning to extend reasoning depth during inference, achieving superior performance over fixed-token baselines.
Langue du texte source : anglais
Enable dynamic control over reasoning depth during inference using Reasoning Control Fields that guide long chain-of-thought generation based on execution and quality parameters.
Langue du texte source : anglais
Joint optimization of policy and reward models in LLM reinforcement learning by leveraging rule-based reward precision and dynamically constructing training pairs to prevent reward hacking and improve performance.
Langue du texte source : anglais
Enhance spatial reasoning in multimodal LLMs by integrating depth and segmentation as auxiliary modalities with adaptive reasoning strategies. COOPER achieves 6.91% improvement in spatial understanding—when you need 3D-aware vision-language capabilities.
Langue du texte source : anglais
Implement techniques from CooperBench: Why Coding Agents Cannot be Your Teammates Yet. Resolving team conflicts requires not only task-specific competence, but also social intelligence to find common ground and build consensus
Langue du texte source : anglais
Enhance reasoning models by integrating executable code within thinking traces, enabling grounded computation verification and reducing hallucination in mathematical and logical reasoning.
Langue du texte source : anglais