Menu
Skills neste repositório
O SkillsMP coletou 1.228 skills de ADu2021/skillXiv. Abra uma skill para revisar a origem e os detalhes.
ADu2021/skillXivMostrando 40 de 1.228 skills coletadas.
Stabilize RL training on reasoning tasks by performing entropy-guided rollouts from uncertain decision points, avoiding policy collapse and premature convergence. Increases fully correct trajectories on math reasoning while maintaining stable entropy…
Idioma do texto original: inglês
Monitor Shannon entropy in LLM token distributions to detect sudden uncertainty spikes, then trigger adaptive prompt consolidation to realign conversation context and improve accuracy by 56.6% and reliability by 35.3%.
Idioma do texto original: inglês
Stabilize LLM post-training by constraining global distributional shifts in policy exploration. Entropy Ratio Clipping supplements local clipping mechanisms with global entropy constraints—essential when PPO alone produces unstable gradients and distribution…
Idioma do texto original: inglês
Automatically generate diverse, scalable tool-interactive training environments for LLM agents without manual sandbox creation. Uses topic mining and logic modeling to create varied environment architectures with task scenarios, enabling agents to learn…
Idioma do texto original: inglês
Systematic evaluation toolkit for assessing large language models across multiple dimensions, enabling comprehensive benchmarking of agent capabilities and comparative analysis of model performance.
Idioma do texto original: inglês
Stabilize multi-turn LLM agent training with entropy-regularized policy optimization that prevents exploration-exploitation cascade failures in sparse-reward environments through trajectory-level entropy regulation, historical smoothing, and adaptive…
Idioma do texto original: inglês
Transform vision-language models into embodied agents through two stages: learning embodied priors from trajectory-augmented data with LLM reasoning, then online RL with self-summarization and dense rewards for long-horizon tasks.
Idioma do texto original: inglês
24-trillion-token web dataset with 12-category taxonomy labels enabling efficient curation of specialized datasets through SQL filtering and small annotation models.
Idioma do texto original: inglês
Chunk-based landmark-guided sparse attention enabling 16M-token context windows with 90%+ retrieval accuracy on in-context tasks. Use when processing ultra-long documents where full attention is prohibitive but context selection must be dynamic.
Idioma do texto original: inglês
Train autonomous agents to use computers by generating synthetic task experiences and iterating on them, achieving 56.7% success on OSWorld benchmarks through scalable experience-driven optimization. Use when you need agents that autonomously learn complex…
Idioma do texto original: inglês
Scale Evolution Strategies to billion-parameter LLMs without backpropagation for superior robustness and stability across diverse models, reward horizons, and evaluation tasks. Outperforms RL methods while eliminating gradient computation overhead.
Idioma do texto original: inglês
Generate verifiable synthetic training data (problems + solutions + verification artifacts) through evolutionary synthesis that works across math, code, and agent tasks without task-specific rules.
Idioma do texto original: inglês
System for evolving and composing programmatic skills through learning, enabling agents to discover and develop new behavioral capabilities dynamically.
Idioma do texto original: inglês
Build unified LLMs supporting both fast standard inference and slow reasoning modes through hybrid attention and adaptive post-training. Combine non-reasoning and reasoning capabilities in a single model with 128K context windows and tool use. Use when you…
Idioma do texto original: inglês
Build robust reward models for computer-use agents by processing execution videos with spatiotemporal token pruning (STP+TTP) to remove redundancy while preserving task-relevant UI details. Achieve 84.7% accuracy with adversarial instruction translation.
Idioma do texto original: inglês
Improve RLVR training efficiency by selectively replaying trajectories based on correctness and entropy. Medium-difficulty questions and low-entropy solutions are most valuable; selective replay yields +3.5-7.6% improvements.
Idioma do texto original: inglês
Improve LLM reasoning efficiency by systematically reusing past rollouts through experience replay. ExGRPO organizes training data by success and diversity, applying a mixed-policy objective that prioritizes high-quality examples while maintaining…
Idioma do texto original: inglês
Evaluate AI systems' ability to conduct autonomous research experiments using EXP-Bench, a benchmark for multi-step scientific reasoning and iterative experimental workflows.
Idioma do texto original: inglês
Dynamically adapt LLM reasoning strategies at inference time by curating episodic memory of past problem solutions—generate task-specific prompts, tool configs, and control logic for up to 111× cost reduction and 14% accuracy gains.
Idioma do texto original: inglês
Improve RL training efficiency by embedding explicit reflection and revision loops. Models generate initial responses, receive feedback, produce self-reflections describing improvements, revise their attempts, and distill successful corrections into the base…
Idioma do texto original: inglês
Improve MoE language model efficiency with causal threshold-based routing that eliminates auxiliary losses and enables dynamic per-token computation.
Idioma do texto original: inglês
Investigate exploration-exploitation trade-offs in reinforcement learning with verifiable rewards through theoretical analysis and empirical validation. Derive explicit clipping bias bounds, establish policy-entropy shift formulation, and introduce…
Idioma do texto original: inglês
Multimodal system for joint audio-video understanding and generation, enhancing agent perception and communication in environments with both audio and visual information.
Idioma do texto original: inglês
Unify LLM alignment methods through f-divergence theory. f-GRPO extends GRPO to handle any divergence measure (KL, Jensen-Shannon, Hellinger), enabling tailored alignment objectives. f-HAL combines on-policy and off-policy preference learning to prevent…
Idioma do texto original: inglês
Prevent policy collapse onto common solutions during reinforcement learning by applying difficulty-aware focal weighting to gradient contributions, maintaining diversity across solution modes while preserving performance.
Idioma do texto original: inglês
F1 is a 4.2B-parameter Vision-Language-Action model that reformulates robotic control as foresight-guided inverse dynamics. Rather than reactive state-to-action mappings, F1 predicts plausible future visual states and derives actions to achieve them, enabling…
Idioma do texto original: inglês
Enable extreme 1-2bit quantization of pre-trained LLMs by leveraging complex-valued arithmetic. Convert real-valued linear layers to complex domain losslessly, quantize to fourth roots of unity {±1, ±i}, and apply recursive residual error quantization for…
Idioma do texto original: inglês
Achieving human-level performance in Vision-and-Language Navigation (VLN) requires an embodied agent to jointly understand multimodal instructions and visual-spatial context while reasoning over long action sequences. Recent works, such as NavCoT and…
Idioma do texto original: inglês
Generates high-quality images directly from pixels using flow-matching-based latent sequences. Transforms images via invertible flows into manageable latent sequences, applies autoregressive modeling, and uses classifier-free guidance. Provides exact…
Idioma do texto original: inglês
Identify dominant frequency components in RoPE embeddings to determine which attention dimensions are essential, reducing KV cache by 81% while maintaining performance. Training-free approach integrates with existing compression methods for 2.56x speedup on…
Idioma do texto original: inglês
Implements Fast-ThinkAct from arXiv:2601.09708
Idioma do texto original: inglês
Identify and measure feedback friction in LLM reasoning tasks where models resist high-quality guidance, discovering that confidence predicts feedback receptiveness and revealing mitigation strategies.
Idioma do texto original: inglês
Systematically adapt state-of-the-art distillation methods for T2I generation. Compare sCM (stabilized Consistency Models), MeanFlow, and IMM within unified framework. sCM excels at extreme few-step regimes (52.81% GenEval at 2 steps), MeanFlow achieves…
Idioma do texto original: inglês
Demonstrate that adversarial attacks on vision-language models need not target all tokens equally. Entropy-guided attacks identify high-entropy tokens (critical decision points) where perturbations have maximum impact, achieving comparable attack success with…
Idioma do texto original: inglês
Research contribution advancing agent and reasoning capabilities through novel approaches to model development, training, and evaluation.
Idioma do texto original: inglês
Active learning (AL) has the potential to drastically reduce annotation costs in 3D biomedical image segmentation, where expert labeling of volumetric data is both time-consuming and expensive. Yet, existing AL methods are unable to consistently outperform…
Idioma do texto original: inglês
Provide continuous confidence scores throughout LLM text generation via supervised learning and backward confidence integration, enabling real-time uncertainty awareness
Idioma do texto original: inglês
Financial agents powered by large language models (LLMs) are increasingly deployed for investment analysis, risk assessment, and automated decision-making, where their abilities to plan, invoke tools, and manipulate mutable state introduce new security risks…
Idioma do texto original: inglês
Enable video generation customization via first-frame reuse—treat initial frame as a visual entity buffer storing objects for reuse during generation with just 20-50 examples, requiring minimal architectural change.
Idioma do texto original: inglês
Train small models to recover from tool execution errors by treating errors as training signals. Use error simulators to generate recovery examples and fission failed trajectories into multiple recovery attempts.
Idioma do texto original: inglês