adu2021
- 1.2k skills
- 0 followers
- 1 day ago last updated
- ▌ Webworld · adu2021Train autoregressive world models on 1M+ real web interactions for accurate browser state prediction. Enables agent training with 100× more data than prior approaches, achieving GPT-4o comparable performance with format flexibility and cross-domain generalization.
- ▌ R Horizon Long Horizon Reasoning · adu2021Construct 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.
- ▌ Robust R1 · adu2021Enable MLLMs to handle visually degraded images by explicitly reasoning about degradation types and severity. Trains models to perceive degradation parameters, analyze semantic impact, and reconstruct interpretations while adapting reasoning depth to degradation complexity—achieving robust understanding with interpretable explanations.
- ▌ Paper2skill Evaluation Infrastructure · adu2021Convert dataset and benchmark papers into evaluation infrastructure skills. For datasets: extracts collection protocol, annotation design, quality control. For benchmarks: extracts task definition, metric selection, leaderboard design. Use this skill when extracting skills from Category 2 (Evaluation Infrastructure) papers — ImageNet-style dataset papers, SWE-bench-style benchmark papers, or any paper whose primary contribution is evaluation methodology.
- ▌ Deep Prune · adu2021Reduce redundant tokens in parallel reasoning by 80% while maintaining accuracy via dynamic pruning of equivalent reasoning paths. Trigger: improve efficiency of consensus-based reasoning (multiple CoT generation).
- ▌ Nemotron 3 · adu2021Build efficient open-source LLMs via hybrid Mamba-Transformer MoE architecture with LatentMoE expert design, multi-token prediction training, FP4 precision, and multi-environment RL post-training—achieving 3.3× higher throughput than equivalently-sized models while maintaining state-of-the-art reasoning, coding, and tool-use capabilities.
- ▌ Rise Robot · adu2021Enable robot policies to self-improve through imagination using learned dynamics and value models without physical trial-and-error. Compositional world model separates concerns enabling 35-45% performance gains on contact-rich manipulation.
- ▌ Stable Drl · adu2021Fixes training instability in Group Relative Policy Optimization for discrete language models by replacing conditional clipping with strict importance ratio bounds and self-normalized advantages. Prevents gradient spikes and policy collapse.
- ▌ Code2worlds · adu2021Generate 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.
- ▌ Steer2adapt · adu2021Adapt LLMs efficiently by composing multiple pre-existing semantic steering vectors via Bayesian optimization, balancing adaptation gain and safety without retraining model parameters.
- ▌ World Cache · adu2021Accelerates iterative world model inference by classifying tokens by temporal curvature (predictability) and applying differentiated caching: stable tokens reused, linear tokens extrapolated, chaotic tokens updated. Achieves 3.7x speedup with 98% rollout quality.
- ▌ Causal Armor · adu2021Defend against indirect prompt injection attacks by detecting dominance shifts using leave-one-out attribution, enabling selective sanitization without sacrificing latency or utility.
- ▌ Himap Travel · adu2021Solves long-horizon planning problems with global constraints by decoupling planning into strategic (resource allocation) and tactical (execution) levels. Prevents constraint drift through synchronized state tracking and cooperative bargaining.
- ▌ Minicpm Sala · adu2021Combine sparse attention (25% of layers) and linear attention (75% of layers) via strategic layer placement to handle 1M-token contexts with 75% training cost reduction. Hybrid positional encoding preserves long-range information while maintaining position awareness.
- ▌ Simko Pass K · adu2021Improve pass@K by using asymmetric probability boosting: increase probabilities of top-K correct solutions while penalizing top-1 incorrect predictions. Focus boosting on high-entropy tokens where exploration helps most.
- ▌ Browser Agent · adu2021Build web agents using human-inspired browser actions (scrolling, clicking, typing) operated directly on raw HTML via Playwright. Combine supervised fine-tuning and rejection fine-tuning with explicit memory for strong generalization on web tasks.
- ▌ Config Agents · adu2021Learn 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 improvement with 35% cost reduction.
- ▌ Flash Prefill · adu2021Accelerates long-context LLM prefilling by identifying sparse attention patterns without expensive scoring, using block-level approximations and dynamic thresholding. Achieves 27.78x speedup at 256K tokens while maintaining accuracy.
- ▌ Gtr Turbo Vlm · adu2021Eliminate expensive external teacher dependencies in VLM RL training via merged-checkpoint teachers. Uses TIES merging of historical RL checkpoints to create free, stable teacher models for step-level guidance—matching external teacher performance while reducing training time 50% and computational costs 60%.
- ▌ Lamer Meta Rl · adu2021Enable LLM agents to actively explore and adapt policies through meta-RL instead of converging to fixed behaviors. Uses cross-episode training with trajectory-level discounting and in-context policy adaptation via textual reflections—achieving 11-19% improvements in exploration-exploitation balance across interactive environments.
- ▌ Mai UI Agents · adu2021Scale GUI agents to real-world complexity via extended action space (user interaction, tool calls) and device-cloud collaboration. Online RL supports 500+ parallel environments with asynchronous handling; local agent monitors trajectory alignment and handoffs to cloud when drift detected—achieving 41.7% MobileWorld success with privacy-preserving delegation.
- ▌ Reflexi Coder · adu2021Trains LLMs to autonomously debug and improve code through structured RL-optimized reflection cycles. Internalizes debugging process into model weights rather than relying on external oracles or expensive iterative prompting.
- ▌ Sage Rl Agent · adu2021Enable agents to continuously improve by accumulating reusable skills across sequential task chains. Train via GRPO across task sequences where skills persist and compound, provide dual rewards for both task completion and skill generation/reuse—improving completion rates 8.9% while reducing token costs by 59% compared to non-skill baselines.
- ▌ Self Eval T2i · adu2021Train text-to-image models from scratch for any inference step count via self-evaluation mechanism. Model evaluates its own generated samples using current score estimates as dynamic self-teacher, enabling global distribution matching without external teachers—achieving few-step quality equivalent to many-step models at all budgets.
- ▌ Truncated Ppo · adu2021T-PPO improves training efficiency via truncated rollouts and extended GAE, enabling batch continuity without waiting for full sequence completion.
- ▌ Spatial Boost · adu2021Inject spatial understanding into VLMs via language-guided multi-turn Chain-of-Thought reasoning over hierarchical spatial knowledge (pixel→object→scene). Implement dual-channel attention mechanism preserving pre-trained vision features while progressively incorporating dense 3D spatial information. Frozen original parameters prevent catastrophic forgetting; only new channel and mixture weights update. Maintains visual fidelity while enabling precise spatial reasoning (depth, relative positions, distances).
- ▌ AI Coscientist · adu2021Train LLMs to generate high-quality research plans via rubric-based RL without requiring experimental verification. Extracts research goals and domain-specific rubrics from scientific papers, uses frozen model as grader with 12-22% relative improvements, achieves human-expert preference 70% of time with strong cross-domain generalization.
- ▌ Calm Reasoning · adu2021Adapt large reasoning models for optimization tasks using expert-guided hint correction. Generate high-quality training data with minimal expert intervention (<2.6% token modification). Trigger: fine-tune reasoning models on domain-specific tasks without large supervised datasets.
- ▌ Casa Vl Fusion · adu2021Replace token-insertion for fusing vision and language with efficient cross-attention that maintains separate text self-attention. Enables text tokens to attend images within local windows, preserves gist tokens from prior images, and maintains near-constant memory costs for streaming video—more practical than direct token insertion for resource-constrained applications.
- ▌ Eaglet Planner · adu2021Train efficient planners for long-horizon agent tasks using homologous consensus filtering to generate synthetic plans from strong LLMs and rule-based RL with executor capability rewards. Reduces training cost by 8x while maintaining state-of-the-art performance.
- ▌ Lopa Lookahead · adu2021Maximize parallelism in diffusion language models by intelligently exploring token filling orders. Spawn multiple candidate branches predicting high-confidence positions, evaluate all branches in one forward pass, and select the branch enabling most future parallelism—increasing tokens-per-forward-pass 4.4× without accuracy loss.
- ▌ Nabla Reasoner · adu2021Improves LLM reasoning quality at inference time by optimizing token logits using gradient descent, combining reward model signals with KL-regularization. Bridges parametric training-time and non-parametric test-time scaling through token-level optimization.
- ▌ Slm Agentic AI · adu2021Design heterogeneous agentic systems combining specialized small models with selective large model deployment for superior economics and performance.
- ▌ State Thoughts · adu2021Guide multi-step reasoning through structured action templates that decompose problems into discrete, executable steps. STATe improves exploration efficiency in tree-of-thoughts by constraining action space while maintaining reasoning flexibility.
- ▌ Va Pi Pixel Ar · adu2021Align autoregressive image models with pixel-space quality via variational optimization. Formulates alignment as ELBO combining reconstruction (pixel supervision) and prior regularization (token distribution), treating model as RL policy with tokenizer reconstruction as reward—achieving 86.6% cost reduction vs standard RL fine-tuning.
- ▌ Agentcpm Report · adu2021Enable research agents to interleave evidence-based drafting with reasoning-driven deepening, automatically expanding outlines based on discovered gaps, using trajectory pruning for efficient RL training.
- ▌ Alignment Waltz · adu2021Train safety-aligned agents using collaborative multi-agent RL where conversation and feedback agents improve together. Trigger: reduce overrefusal while maintaining safety on sensitive queries.
- ▌ Chain Of Agents · adu2021Train single models to simulate multi-agent collaboration through distillation from complex multi-agent systems and agentic RL, creating efficient Agent Foundation Models for tool use and web navigation.
- ▌ Coco Text Image · adu2021Generates 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.
- ▌ Deep Search Hmc · adu2021Monitor search agent reasoning quality via hierarchical uncertainty detection. Fast consistency checks identify anomalies; slow experience-driven feedback provides corrections. Minimal overhead while catching misalignment.
- ▌ Nemotron 3 Nano · adu2021Efficient agentic reasoning via sparse MoE activating 50% parameters per token. Combines Mamba-Transformer hybrid with 6-of-128 expert routing, three-stage post-training (SFT, verifiable RL, RLHF), and Group Relative Length Control—achieving 3.3× inference throughput of competitors while maintaining 1M token context support and superior reasoning.
- ▌ Qwen3 Embedding · adu2021Leverages 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.
- ▌ Seed Prover 1 5 · adu2021Enable LLM agents to incrementally construct formal proofs through strategic tool orchestration rather than one-shot generation. Combines natural language reasoning, sketch translation, and formal verification in an agentic workflow trained via PPO and Rubric RL, achieving competitive performance on undergraduate and graduate proof problems.
- ▌ Srpo Vla Policy · adu2021Train VLA models for robotic manipulation by using the model's own successful trajectories as self-reference for reward—enable progress-based feedback for failed attempts without external rewards or demonstrations.
- ▌ Swe Rm Verifier · adu2021Improve SWE agent RL via execution-free reward models optimized for three metrics beyond TTS: discriminative ability (AUC) and calibration (ECE). Shows TTS alone insufficient—models must distinguish correct/incorrect trajectories and align confidence with actual correctness—improving SWE-Bench Verified by 7-10 points with 30B MoE verifier.
- ▌
- ▌ Vlash Async Vla · adu2021Rolls forward robot state using previously generated actions to condition predictions on estimated future states, paired with temporal-offset training augmentation. Achieve 2× speedup and 17.4× latency reduction in vision-language-action models without architectural changes.
- ▌ Open Researcher · adu2021Replace live web API-dependent research with offline corpus-based trajectory synthesis. Decouples answer-guided document retrieval (10K gold + 15M FineWeb) from synthesis via local search engine, eliminating $5,760 Serper costs while enabling reproducible, analyzable reasoning chains through three primitives: Search (ranked retrieval), Open (full document fetch), Find (intra-document verification).
- ▌ Video Detective · adu2021Enable VLMs to find relevant clips in long videos through sparse observation and graph-based propagation. Iteratively hypothesize promising segments, extract multimodal evidence (captions, OCR, speech), and propagate relevance scores via visual-temporal affinity graph. Jointly model extrinsic relevance (query-to-segment matching) and intrinsic relevance (video internal structure) to infer unobserved segments. Achieves competitive accuracy with significantly fewer tokens than dense sampling.
- ▌ Bottom Up Policy · adu2021Optimize language model policies layer-by-layer rather than monolithically to understand internal reasoning structure. Decompose models into per-layer and per-module policies via residual streams, analyze entropy patterns revealing exploration→convergence phases, and optimize layers sequentially—improving reasoning on math tasks by up to 4.69 points.
- ▌ Complementary Rl · adu2021Improve RL sample efficiency through co-evolution of policy and experience extractor, enabling dynamic experience replay that adapts to the agent's skill level.
- ▌ Dmtd Multi Token · adu2021Generate 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.
- ▌ Memory As Action · adu2021Treat memory management as learnable RL policy actions (delete/insert) rather than fixed mechanisms. Enable models to autonomously decide what to keep, remove, or add to context, reducing average context length by 51% while matching larger models.
- ▌ Prism Hypothesis · adu2021Unify 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.
- ▌ Robo Dopamine Rl · adu2021Overcome reward function design challenges via General Reward Model (GRM) for step-wise progress assessment. Uses multi-view observations for occlusion robustness, hop-based progress normalization, Policy-Invariant Reward Shaping—enabling 95% robot task success within 150 interactions with theoretically-grounded dense rewards.
- ▌ Shape Of Thought · adu2021Demonstrate that synthetic CoT traces with incorrect final answers outperform human-written correct solutions for supervised fine-tuning. Distribution proximity between training data and student model's natural output matters more than correctness—validating human traces with model-like distributions improves performance, providing practical guidance for dataset curation.
- ▌ Smartsnap Agents · adu2021Shift agent verification from post-hoc external judgment to proactive in-situ self-evidence curation. Agents generate atomic evidence tuples during execution, guided by 3C principles (Completeness, Conciseness, Creativity), with structured verifier feedback across four dimensions—reducing verification costs and enabling dense learning signals.
- ▌ Test Time Rl Vlm · adu2021Adapt vision-language models at inference without labeled data by generating multiple predictions and rewarding high-frequency outputs. Trigger: improve VLM accuracy on deployment with self-generated supervision signals.
- ▌ Tool Learning Lm · adu2021Prove that tool-augmented learning unboundedly scales factual knowledge recall compared to parameter-constrained memorization, enabling efficient knowledge retrieval via external tools
- ▌ Web World Models · adu2021Bridge deterministic web code and generative LLMs via hybrid world models. State and physics defined by TypeScript code, LLMs generate narrative/aesthetics on top. Uses typed interfaces, deterministic hashing, graceful degradation—enabling scalable interactive environments from travel atlases to fictional worlds without databases.
- ▌ X Vla Embodiment · adu2021Use soft-prompted transformer architecture for multi-robot learning. Add learnable embodiment-specific prompt embeddings to handle different robot types while maintaining single shared backbone. Scale to 0.9B parameters across 6 simulators and 3 real robots.
- ▌ Omni World Bench · adu2021Evaluate world models on faithful interactive response capability through three hierarchical interaction levels (single object, localized, multi-object effects) and four complementary metrics measuring video quality, controllability, and physical plausibility. Identifies the gap between visual fidelity and true interactive state transition modeling.
- ▌ Confu Speculative · adu2021Improves 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.
- ▌ Dirl Diffusion Rl · adu2021Enable 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 performance to rival larger autoregressive models.
- ▌
- ▌
- ▌ Finece Confidence · adu2021Provide continuous confidence scores throughout LLM text generation via supervised learning and backward confidence integration, enabling real-time uncertainty awareness
- ▌ Laser Self Reward · adu2021Compute reasoning rewards from the model's own next-token probability distribution at solution end. Integrates verification-based feedback into single model without separate evaluator, enabling efficient RL training with minimal overhead.
- ▌ Long Cot Training · adu2021Navigate Long CoT Degradation phenomenon when training small models on extended reasoning, understanding recovery dynamics and implementing strategies to maintain performance.
- ▌ M3 Bench Tool Use · adu2021Design and evaluate multimodal agents for tool use with M3-Bench: assess three interconnected dimensions (multi-modal grounding, multi-hop causality, multi-threaded parallelism) using similarity-bucketed Hungarian alignment for transparent tool call evaluation without LLM judges.
- ▌ Markovian Thinker · adu2021Enable LLMs to scale reasoning length from O(n²) to O(n) by structuring thinking into fixed-size chunks with learnable cross-chunk summaries. Trigger: train reasoning models with unbounded or expensive chain-of-thought sequences.
- ▌ Qerl Quantization Rl · adu2021Combine 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.
- ▌ R Wom World Model · adu2021Ground 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.
- ▌ Reasoning Palette · adu2021Overcome token-level randomness limitations in RL by shifting exploration to latent reasoning strategies. Train a VAE encoding diverse reasoning patterns, sample latents during RL, decode to prefix embeddings steering internal reasoning—enabling structured exploration across math, coding, and QA with interpretable, controllable behavior.
- ▌ Rl Rubric Anchors · adu2021Extend RL to open-ended tasks using structured rubrics as reward anchors, enabling fine-grained evaluation of subjective outputs without requiring binary correctness signals.
- ▌ Scaling Laws Code · adu2021Establish language-specific scaling laws for multilingual code pre-training across 7 programming languages (0.2B-14B models, 1T tokens). Show interpreted languages scale better than compiled, synergy gains depend on syntax similarity, and parallel pairing improves translation—enabling proportion-dependent token allocation outperforming uniform distribution.
- ▌ Semanticgen Video · adu2021Accelerate video generation and enable long-video synthesis by decomposing into two diffusion stages: first generate compact semantic features for global planning, then generate VAE latents conditioned on semantics. Includes learnable semantic compression to improve training convergence—enabling minute-long videos with faster convergence than direct VAE modeling.
- ▌ Step Deepresearch · adu2021Enable autonomous AI research systems to move beyond web search toward true research through four composable atomic capabilities: planning/decomposition, deep search/synthesis, reflection/verification, and report generation. Progressive training across 32K→128K context with SFT and RL produces single ReAct agents matching proprietary systems with lower costs.
- ▌ Green Vla Robotics · adu2021Train robot controllers via five-stage curriculum progressing from base vision-language models to embodiment-specific RL-refined policies. Unified action space enables cross-embodiment transfer with minimal performance loss.
- ▌ Ieap Image Editing · adu2021Enable robust image editing by decomposing free-form instructions into sequential atomic operations executed through a neural program interpreter.
- ▌ Llds Grpo Collapse · adu2021Identifies Lazy Likelihood Displacement (LLD) as GRPO failure mechanism in tool-integrated LLMs and proposes lightweight LLDS regularization that penalizes only tokens responsible for likelihood decrease, achieving up to 45.2% performance gains.
- ▌ Lynx Dynamic Exits · adu2021Enable models to stop generating when confident through lightweight hidden-state probes with distributional guarantees. LYNX achieves cross-domain transferability without retraining—ideal when you need confidence-controlled reasoning efficiency.
- ▌ Memory T1 Temporal · adu2021Enable agents to accurately identify temporally relevant information in long multi-session dialogues through RL-based memory retrieval. Combines coarse-to-fine candidate selection with multi-level temporal consistency rewards—providing dense supervision that disambiguates time expressions and maintains coherence across 128k-token contexts.
- ▌ Mm Helix Reasoning · adu2021Train multimodal models for long-chain reflective reasoning (iterative thinking, backtracking) using Adaptive Hybrid Policy Optimization. Trigger: improve VLM performance on tasks requiring iterative refinement and error correction.
- ▌ Multi Agent Evolve · adu2021Enables LLM self-improvement without external verification through multi-agent co-evolution. Proposer generates questions, Solver attempts solutions, Judge evaluates both. All agents evolve together via RL, achieving 4.54% improvement on reasoning benchmarks without human supervision.
- ▌ Pass At K Training · adu2021Use Pass@k as the reward metric in RL training to balance exploration and exploitation, enabling models to learn diverse sampling strategies while maintaining correctness.
- ▌ Play To Generalize · adu2021Develop reasoning capabilities in multimodal models through interactive gameplay rather than supervised learning, achieving superior transfer to diverse reasoning tasks.
- ▌ Remix Lora Routing · adu2021Learn to route requests across multiple LoRA adapters using RL-based router training with constant routing weights. Prevents weight collapse and ensures balanced contribution from all selected adapters during inference.
- ▌ Self Forcing Video · adu2021Address exposure bias in video generation by training on self-generated sequences, achieving 17 FPS real-time performance while matching quality of slower baselines.
- ▌ Self Rewarding Vlm · adu2021Enable VLMs to self-assess visual perception accuracy through decomposed two-stage reasoning: perception generation then validation, eliminating external supervision dependency
- ▌ Single Matrix Lora · adu2021Replace LoRA's two-matrix decomposition with a single learnable matrix (AA⊤) to eliminate scale imbalances and improve training stability. Reduces parameters by ~50% while maintaining or exceeding LoRA performance.
- ▌ Thinking To Recall · adu2021Demonstrates that chain-of-thought reasoning improves LLM factual retrieval through computational buffering and self-priming. Improves single-hop factual accuracy by enabling models to generate contextual bridge facts before recalling answers.
- ▌ Timebill Inference · adu2021Guarantee LLM inference completes within time budgets via response length prediction, execution time estimation, and adaptive KV cache eviction. Three-stage pipeline predicts response length, estimates end-to-end time with 1.22% accuracy, adjusts cache eviction ratio—enabling time-critical deployment in robotics and autonomous systems.
- ▌ Training Free Grpo · adu2021Improve LLM outputs without parameter updates using learned token priors that guide inference. Trigger: optimize agent behavior in deployment without model retraining or fine-tuning.
- ▌ Yolo Moe Detection · adu2021MoE-enhanced YOLOv9-Tiny using lightweight routers to adaptively weight specialized expert outputs at each feature resolution, improving detection quality while maintaining end-to-end differentiability. Deploy for efficient multi-scale object detection with dynamic specialization.
- ▌ Agent Data Protocol · adu2021Standardizes agent training data representation across diverse sources (API use, web browsing, coding, software engineering). Single lightweight protocol unifies 13 datasets enabling 20% performance gains without domain-specific tuning. Enables reproducible agent training and scalable data combination.
- ▌ Cort Code Reasoning · adu2021Enhance reasoning models by integrating executable code within thinking traces, enabling grounded computation verification and reducing hallucination in mathematical and logical reasoning.
- ▌ Era Embodied Agents · adu2021Transform 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.
- ▌ Gui Actor Grounding · adu2021Enable GUI agents to ground actions without generating pixel coordinates by using attention-based patch-level alignment and a verifier for selecting optimal action regions from candidates.
- ▌ Kernelevole Agentic · adu2021Automate compute kernel optimization via agentic AI and retrieval-augmented prompting. Unified context-aware transformation function, hardware-specific constraint KB, self-improving state machine exploring kernel variants—achieving 1.25-17× speedups on production recommendation workloads in hours vs weeks of manual effort.