adu2021
- 1.2k skills
- 0 followers
- 1 day ago last updated
- ▌ Dcm Dual Expert Consistency · adu2021Accelerate video generation through dual-expert consistency distillation, using separate denoisers for semantic layout/motion and detail refinement to resolve conflicting optimization gradients.
- ▌ Decoupled Reasoning Proving · adu2021Bridge the gap between informal mathematical reasoning (80% accuracy) and formal proof synthesis (8% success) by decoupling them: a general-purpose reasoner generates strategic lemmas, then a specialized prover verifies them formally. First open-source solver of post-2000 IMO problems.
- ▌ Diffusionvl Ar To Diffusion · adu2021Convert 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, causal between blocks. Requires less than 5% of data compared to prior diffusion VLM methods.
- ▌ Dive Diverse Task Synthesis · adu2021Synthesize 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.
- ▌ Driving Gen Video Benchmark · adu2021DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving. From arXiv:2601.01528
- ▌ Dvoting Diffusion Consensus · adu2021Accelerate test-time scaling for diffusion language models by identifying inconsistent tokens, selectively remask and regenerate only uncertain tokens, and aggregate across samples via voting. Achieve 5.5-22× speedup over standard iterative sampling with 6-8% accuracy gains on reasoning tasks.
- ▌ Efficient Dlm Ar Conversion · adu2021Systematically convert pretrained autoregressive models into efficient diffusion language models via block-wise attention and position-dependent masking. Efficient-DLM family (1.5B/4B/8B) maintains comparable accuracy to standard AR models while delivering 4.5× higher throughput.
- ▌ Entropy Guided Regeneration · adu2021Monitor 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%.
- ▌ Evolutionary Data Synthesis · adu2021Generate verifiable synthetic training data (problems + solutions + verification artifacts) through evolutionary synthesis that works across math, code, and agent tasks without task-specific rules.
- ▌ F Grpo Divergence Alignment · adu2021Unify 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 reward hacking while maintaining safety alignment.
- ▌ Flow Map Trajectory Tilting · adu2021Uses flow maps as look-ahead operators to enable principled reward-guided diffusion by predicting trajectory endpoints at any denoising step. Deploy when applying rewards or preferences to diffusion trajectories with meaningful gradients throughout generation.
- ▌ Gain Rl Angle Concentration · adu2021Improve RL training efficiency by 2.5× using angle concentration between token hidden states as a cost-effective data scheduling signal, selecting high-gradient samples dynamically.
- ▌ Gem Agentic LLM Environment · adu2021A standardized environment framework for training and evaluating LLM agents, providing 24+ tasks with asynchronous vectorized execution, extensible wrappers, and integration examples for five RL frameworks. Enables reproducible agent research and training at scale.
- ▌ Golden Goose Task Synthesis · adu2021Synthesize unlimited verifiable training tasks from unverifiable text by converting reasoning passages into multiple-choice problems. Creates higher-quality training data for RLHF systems without requiring new human labels.
- ▌ Hardtests Code Verification · adu2021Generate comprehensive test cases for code problems that reliably detect wrong solutions through LLM-based edge case synthesis and test quality ranking.
- ▌ Kling Omni Multimodal Video · adu2021Unify video generation, editing, and reasoning through Multimodal Visual Language (MVL) paradigm processing text instructions, reference images, and video contexts in shared embedding space. Employ prompt enhancer, omni-generator with diffusion transformer, and multimodal super-resolution. Support diverse user inputs with in-context generation and reasoning-based editing.
- ▌ Kv Cache Steering Reasoning · adu2021Guide frozen language models toward multi-step reasoning by modifying cached key-value representations after the prefilling stage. Extract steering vectors from contrastive prompt pairs and apply them to KV cache with scalar coefficients. Improves reasoning on GSM8K, ARC, CommonsenseQA while adding only 10ms overhead per token.
- ▌ Learning Discover Test Time · adu2021Apply reinforcement learning at test time to enable language models to continue adapting on specific problems, achieving state-of-the-art results across mathematics, GPU optimization, algorithms, and biology. Use when you need models to discover domain-specific problem-solving strategies without retraining.
- ▌ Log Linear Sparse Attention · adu2021Reduce self-attention complexity from O(N²) to O(N log N) through hierarchical token selection and enrichment. Perform hierarchical Top-K selection progressively adopting sparse Top-K at each level. Implement sparse index transpose algorithm avoiding dense mask construction. Achieve 28.27× faster inference and 6.09× faster training.
- ▌ Longcontext Reward Modeling · adu2021Train reward models to evaluate long-context responses by introducing faithfulness as a criterion and using consistency-based alignment to maintain judgment-explanation coherence.
- ▌ Longllada Diffusion Context · adu2021Training-free method extending diffusion LLMs to 6x context length using NTK-based RoPE scaling, exploiting bidirectional attention stability.
- ▌ Mact Document Understanding · adu2021Decomposes document processing into specialized agents (planning, execution, judgment, answer) with agent-wise adaptive test-time scaling. Achieves 9.9-11.5% performance gain with smaller models while maintaining reasoning.
- ▌ Memento Skills Agent Design · adu2021Enable agents to autonomously design and refine task-specific agents by evolving externalised behavioral skills and prompts without modifying base LLM parameters.
- ▌ Mirror Speculative Decoding · adu2021Run draft and target models in parallel across GPU/NPU using bidirectional speculation: draft predicts forward continuations while target speculates correction paths simultaneously. Achieve 2.8x-5.8x wall-time speedup on 14B-66B models.
- ▌ Mixture Of Depths Attention · adu2021Allow attention heads to reference features from multiple depths by accessing both current-layer and depth key-value pairs. Prevent signal degradation in deep models while maintaining computational efficiency.
- ▌ Molecular Thought Reasoning · adu2021Improve agent reasoning by designing thought structures that balance deep analysis, self-reflection, and exploratory thinking. Framework discovers that effective long-form reasoning exhibits molecular-like interaction patterns—specific bonds between reasoning components that enable fast entropy convergence. Method synthesizes improved reasoning trajectories using distribution-transfer, improving both model performance and RL training stability.
- ▌ Multi Agent Tool Policy Opt · adu2021Train planner and worker agent roles within a single LLM via role-specific prompts and RL, avoiding multi-instance overhead while preserving specialization. Trigger: improve tool-use planning robustness to noisy outputs without deploying separate models.
- ▌ Opensir Self Play Reasoning · adu2021Enable open-ended mathematical learning through co-evolutionary teacher-student self-play where a single model alternates roles to generate progressively harder problems and solve them, creating a virtuous cycle without external data.
- ▌ Paced Weighted Distillation · adu2021Weight distillation loss by student pass rate using Beta kernel: suppress mastered (high pass rate) and intractable (low pass rate) problems, prioritize intermediate zone where learning signal is richest.
- ▌ Paceevolve Evolution Search · adu2021Improves LLM-based evolutionary search by addressing context pollution, mode collapse, and weak collaboration through hierarchical context management, momentum-based backtracking, and adaptive sampling policies.
- ▌ Pretraining Data Refinement · adu2021Improve 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.
- ▌ Prl Process Reward Learning · adu2021Improves 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.
- ▌ Progressive Residual Warmup · adu2021Improves 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 reduction.
- ▌ Prompt Orchestration Markup · adu2021Use 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.
- ▌ Query Bandits Hallucination · adu2021QueryBandits adaptively learns per-query rewriting strategies to reduce LLM hallucinations, achieving 87.5% improvement without model retraining.
- ▌ Reason Rank Passage Ranking · adu2021Enhances passage ranking through reasoning capabilities via synthesized training data and multi-stage training combining supervised fine-tuning with reinforcement learning for improved ranking accuracy.
- ▌ Recall Hierarchical Merging · adu2021Prevents catastrophic forgetting in continual learning by merging models using layer-wise hidden representations as similarity proxies. Shallow layers preserve domain-general features while deep layers enable task-specific adaptation, enabling seamless multi-domain integration without task labels or historical data.
- ▌ Rl Plus Capability Boundary · adu2021Hybrid-policy optimization combining multiple importance sampling and exploration-based advantage functions. Prevents capability boundary collapse while maintaining verification rewards, enabling LLMs to exceed baseline boundaries.
- ▌ Rlad Knowledge Distillation · adu2021RLAD performs selective distillation during RL, using trust-region ratio objectives to balance exploration, exploitation, and teacher guidance for efficient reasoning model training.
- ▌ Robot R1 Embodied Reasoning · adu2021Train vision-language models for robotic manipulation using RL to improve embodied reasoning about spatial relationships and movements, achieving 31% improvement on manipulation benchmarks.
- ▌ Sample More Think Less Gfpo · adu2021Group Filtered Policy Optimization (GFPO) reduces inference-time computation by sampling larger groups during training and filtering responses based on length and token efficiency to teach models efficient reasoning.
- ▌ Segment Policy Optimization · adu2021Segment-level credit assignment for RL in LLMs using Monte Carlo advantage estimation, enabling precise reward attribution without critic models for improved reasoning task performance.
- ▌ Sodec Diffusion Compression · adu2021Replaces iterative diffusion with single-step decoding for image compression. Combines VAE latents with fidelity guidance and rate annealing training. Achieves 20× decoding speedup with improved perceptual quality.
- ▌ Spark Process Aware Rewards · adu2021Train process reward models without ground-truth references using synthetic verification data from generators and verifiers. SPARK achieves 67.5 F1 on ProcessBench—ideal when step-level annotations are expensive but verification is available.
- ▌ State Over Tokens Reasoning · adu2021Reframe reasoning tokens as externalized computational state—the sole persistent information carrier across stateless generation cycles—not human-readable explanations. Model LLM generation as recursive token prediction where state diverges from human semantic interpretation.
- ▌ Static Constrained Decoding · adu2021STATIC converts prefix trees into sparse matrices for vectorized constrained decoding, achieving 948x speedup over CPU and enabling production-scale recommendation systems.
- ▌ Step Gui Calibrated Rewards · adu2021Train GUI agents through calibrated step-level reasoning anchored to trajectory-level validation. Use trajectory verification rather than step annotation achieving 90% accuracy with 10-100× cost reduction. Implement self-evolving training framework with generation-refinement cycles and verifiable reward signals.
- ▌ Stepwiser Generative Judges · adu2021Train stepwise judges as generative models that perform meta-reasoning about intermediate steps, combining explainability with improved accuracy over static process reward models
- ▌ Streambp Efficient Backprop · adu2021Enables 2.8-5.5x longer sequences during LLM training via linear decomposition of chain rule along sequence dimension, maintaining exact gradients with lower memory cost.
- ▌ T3d Trajectory Distillation · adu2021Accelerate diffusion language models from many to few generation steps using trajectory self-distillation. Collect clean-noisy state pairs along teacher trajectories, apply reverse-KL mode-seeking distillation, and weight losses by decoding order to prioritize early predictions where cascading errors compound.
- ▌ Taoavatar Augmented Reality · adu2021Create real-time full-body talking avatars for AR using hybrid parametric-Gaussian representations. Teacher-student distillation transfers pose-dependent deformations from a large network to a compact student model, enabling 90+ FPS rendering on mobile devices.
- ▌ Teso Black Box Optimization · adu2021Apply tabu search enhanced with short-term tabu lists and long-term elite memory for simulation optimization with expensive, noisy evaluations. Balances exploration (avoiding cycling) and exploitation (leveraging best solutions). Use for multimodal landscapes where function evaluations are costly and multiple function calls per solution are impractical.
- ▌ Thinking Map Spatial Agents · adu2021Improve agent reasoning for spatial tasks by augmenting LLMs with map tools and parallel test-time exploration. Framework uses reinforcement learning to train agents to iteratively refine hypotheses using map feedback. Parallel exploration enables agents to test multiple candidate locations before committing to answers. Introduces MAPBench benchmark for evaluating spatial reasoning in image geolocalization tasks.
- ▌ Tool Orchestra Coordination · adu20218B parameter orchestrator trained with end-to-end RL balancing outcome, efficiency, and user preference rewards to strategically coordinate diverse tools and models. Generate realistic tool-use data via ToolScale synthetic pipeline for verifiable multi-turn scenarios.
- ▌ Tool Verification Reasoning · adu2021Tool Verification stabilizes self-improving reasoning models by using external tool execution as ground-truth evidence to prevent spurious consensus from becoming reinforced training signals.
- ▌ Vcrl Variance Curriculum Rl · adu2021VCRL improves LLM mathematical reasoning by dynamically adjusting training sample difficulty based on group reward variance. Uses variance-based sampling with memory replay to focus on moderately-difficult samples where models succeed ~50% of the time, achieving 4.67-point improvement over GRPO on 8B models.
- ▌ Videovla Robot Manipulators · adu2021Transform video generation models into robot manipulators by jointly predicting actions and future visual outcomes. VideoVLA demonstrates that imagining futures improves action reliability—ideal when you need generalizable robot learning from video imagination.
- ▌ Voyager 3d Scene Generation · adu2021Generate spatially-coherent 3D point-cloud videos from single images using depth-fused diffusion with efficient world caching for infinite scene exploration.
- ▌ Zero Shot Motion Generation · adu2021Generate diverse human motions from text descriptions using million-scale datasets and wavelet-enhanced quantization. Achieves state-of-the-art zero-shot generalization on out-of-domain motions, compositional descriptions, and complex choreography through efficient tokenization and scalable transformer-decoder architectures.
- ▌ Demystifying Rl Tool Agents · adu2021Comprehensive recipe for RL-training tool-using agents spanning reward design, data synthesis, model scaling, and algorithm selection. Seven ranked findings: scale-dependent rewards (curriculum for 1.5B–3B; dense for 7B), semi-sparse 'Macro' rewards balance specialization/transfer, 1K-sample sweet spot with 4:3:3 difficulty mix. Achieves SOTA on TravelPlanner with smaller models than leading proprietary systems.
- ▌ Perceptio Spatial Token Vlm · adu2021Enhance VLM spatial grounding by enforcing explicit intermediate spatial reasoning before text generation. Generate segmentation and depth tokens as perception pathways, anchoring responses in 2D and 3D geometric reasoning for improved referring expressions and spatial tasks.
- ▌ Scalable Prompt Routing Moe · adu2021Route queries across frontier models using two-stage system: graph-based task discovery identifies ~332 latent task types via semantic similarity + preference patterns; MoE with task-specific adapters estimates quality for candidate models. Achieves <50% inference cost of strongest single model while exceeding its performance; applies when managing pools of frontier models with narrow capability gaps.
- ▌ Scaling Dora Factored Norms · adu2021Optimize adapter parameter efficiency at scale by decomposing row-wise norm computation into base/cross/BA components (15× memory reduction) and fusing kernel operations. Achieves 1.5–2.0× inference speedup with 77 GB peak VRAM reduction across 8–32B vision-language models; applies when training adapter-based models with strict memory budgets across hundreds of modules.
- ▌ Agent Long Context Benchmark · adu2021Build controllable benchmarks for evaluating long-context agents using environment rollouts. Generate diverse multi-step agent tasks that require maintaining context across extended interaction sequences, enabling evaluation of agent reasoning quality in scenarios with long history requirements.
- ▌ Agent Reasoning Reward Model · adu2021Build multi-faceted reward models for agent trajectories that provide structured feedback on intermediate reasoning quality. Implement explicit reasoning traces, focused critiques with refinement guidance, and overall process scores to train more effective agentic agents without relying solely on sparse outcome rewards.
- ▌ Alphaone Test Time Reasoning · adu2021Dynamically modulate reasoning depth at test time using alpha moments and Bernoulli scheduling to optimize inference speed-quality tradeoffs without retraining.
- ▌ Amft Meta Learning Alignment · adu2021Use meta-learning to automatically balance Supervised Fine-Tuning and Reinforcement Learning signals, treating SFT and RL as complementary rewards in a unified single-stage training framework.
- ▌ Automatic Triton Programming · adu2021Train specialized LLMs to generate optimized Triton GPU kernels using RL with dual rewards for correctness and syntax compliance. 8B model achieves parity with Claude-Sonnet and DeepSeek-R1 by combining supervised fine-tuning on curated code pairs with RL exploration beyond imitation learning ceilings.
- ▌ Bifrost Patch Clip Diffusion · adu2021Connects multimodal language models with diffusion models using patch-level CLIP embeddings as shared latent variables, enabling controllable image generation with minimal training overhead.
- ▌ Bro Rl Broad Rollout Scaling · adu2021Overcome reasoning model training plateaus by increasing rollouts per prompt (N=512) rather than training steps, addressing unsampled coupling that destabilizes learning. Theoretical analysis shows broad exploration eliminates plateau bottleneck.
- ▌ Camels Computer Use Security · adu2021Protects computer use agents from prompt injection by using single-shot execution planning that generates complete control flow graphs before UI observation, preventing instruction hijacking while maintaining 57% performance on frontier models.
- ▌ Comorag Cognitive Memory RAG · adu2021Iteratively retrieve and reason over long narratives using a dynamic memory workspace that integrates retrieved facts into a shared context for complex multi-hop reasoning.
- ▌ Cross Domain Agent Knowledge · adu2021Create a universal memory infrastructure enabling agents across different frameworks to share experience trajectories without retraining. Improve agent performance by retrieving workflows from related domains and applying diagnostic fixes.
- ▌ Deep Research Eval Framework · adu2021DeepResearchEval: An Automated Framework for Deep Research Task Construction and Agentic Evaluation. From arXiv:2601.09688
- ▌ Depth Anything Any Condition · adu2021Extend 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 real-world scenarios.
- ▌ Divide And Conquer Reasoning · adu2021Train 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.
- ▌ Dr Zero Self Evolving Search · adu2021Dr. Zero: Self-Evolving Search Agents without Training Data. From arXiv:2601.07055
- ▌ Efficient Machine Unlearning · adu2021Framework for efficient machine unlearning that reformulates forgetting as inverse learning. Achieves significant computational speedup by replacing expensive Hessian operations with gradient-based optimization, enabling privacy-preserving model updates.
- ▌ Encoder Pretraining Strategy · adu2021Choose optimal pretraining strategy for text encoders: pure MLM, pure CLM, or biphasic CLM-then-MLM training, with empirical guidance on performance across downstream tasks.
- ▌ Exevrm Video Reward Modeling · adu2021Build 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.
- ▌ Flowblending Video Inference · adu2021Accelerate video generation by allocating smaller models to intermediate diffusion timesteps and larger models to capacity-critical early and late stages. Achieves 1.65x speedup and 57% FLOP reduction while maintaining visual quality. Use when video generation latency or computational cost is critical and you have multiple model sizes available.
- ▌ Focus Agent Context Trimming · adu2021Use a lightweight LLM to filter accessibility tree observations by task relevance, reducing agent context size by 50-80% while maintaining equivalent task performance.
- ▌ Env Scaler Synthesis · adu2021Automatically 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 complex multi-turn, multi-tool interactions. Synthesis framework tested on 191 environments with ~7,000 scenarios, improving Qwen3 model performance on knowledge-intensive and search tasks.
- ▌ G2rl Gradient Guided · adu2021Guide LLM exploration through the model's own gradient geometry rather than external signals. Extract sequence-level gradient features measuring how tokens would reshape output distributions. Reward responses introducing novel gradient directions while deemphasizing redundant ones. Achieve orthogonal gradient directions and improved accuracy.
- ▌ Illusion Of Thinking · adu2021Evaluate reasoning model capabilities by analyzing three complexity-dependent behavioral regimes and identifying fundamental limitations in symbolic manipulation rather than computational budgets.
- ▌ Justrl Simple Recipe · adu2021Demonstrate that simple single-stage RL with fixed hyperparameters matches complex multi-stage approaches for training small LLMs on mathematical reasoning. Use basic setup: GRPO algorithm, rule-based verification, 16K token context, standard training data without difficulty filtering. Train two 1.5B models to competitive performance using 2× less compute than sophisticated approaches.
- ▌ Kv Cache Compression · adu2021Enables 8x KV cache compression with minimal training overhead to improve reasoning accuracy by allowing more token generation within computational budgets.
- ▌ Low Prob Exploration · adu2021Preserve reasoning diversity in RL by protecting low-probability tokens that represent novel thinking paths. Trigger: maintain exploration capability during RL training while avoiding entropy collapse on reasoning tasks.
- ▌ Meta Aware Reasoning · adu2021Improve reasoning models by aligning their meta-predictions with actual rollouts through self-generated training signals. Trigger: accelerate reasoning model training while maintaining performance through better meta-cognitive awareness.
- ▌ Omni Reward Modeling · adu2021Builds generalist reward models evaluating outputs across 5 modalities (text, image, video, audio, 3D) using free-form preference data. Combines discriminative and generative reward modeling approaches. Covers 9 tasks with 317K preference examples, enabling preference-based alignment for diverse output modalities.
- ▌ Open Vision Reasoner · adu2021Transfer reasoning behaviors learned in language models to visual domains through two-stage training: cold-start linguistic fine-tuning followed by multimodal RL. Open-Vision-Reasoner achieves 95.3% on MATH500 and 54.6% on MathVerse by learning visual analogs of backtracking, verification, and subgoal decomposition using rule-based rewards.
- ▌ Opv Process Verifier · adu2021Verify long reasoning chains by combining outcome and process verification through active learning. OPV achieves 83.1 F1 on verification—crucial when both final answers and reasoning paths must be validated with minimal annotation.
- ▌ Pear Sft Preparation · adu2021Improve post-RL performance by reweighting SFT loss using importance sampling. Prioritize training examples that match the target policy distribution, not the behavior policy. Achieves 14.6% Pass@8 gains on AIME.
- ▌ Peft Rlvr Evaluation · adu2021Comprehensively evaluate 12+ parameter-efficient fine-tuning methods for RL with Verifiable Rewards (RLVR). Show DoRA/AdaLoRA outperform LoRA, SVD-based methods fail on RL, extreme reduction creates bottlenecks—providing empirical evidence that geometric-aware adapters align better with RL's off-principal update dynamics.
- ▌ Prophet Diffusion Lm · adu2021Detect when diffusion language models converge on correct answers before completing refinement steps using confidence gap monitoring, achieving 3.4x decoding speedup
- ▌ Repro Process Reward · adu2021Dual-scoring mechanism (Magnitude and Stability scores) enabling process-level rewards that penalize inefficient reasoning patterns like overthinking and backtracking without trained reward models. Improve reasoning efficiency in RL without additional supervision.
- ▌ Sage Self Hinting Rl · adu2021Inject privileged hints during GRPO training to reshape rollout distributions when advantage collapses occur, increasing outcome diversity without changing task rewards. Hints removed at deployment; policy automatically learns when to use hints via online refresh mechanism.
- ▌ Self Evolving Agents · adu2021Enable agents to learn continuously from execution experience through hierarchical memory and autonomous reflection. Trigger: improve agent performance on long-horizon tasks by accumulating and applying experience.
- ▌ Smolvla Robotics Vla · adu2021Deploy compact vision-language-action models that run on consumer GPUs for natural language robot control.
- ▌ Superwriter Longform · adu2021Generate coherent, consistent long-form text through structured planning, hierarchical reflection, and Monte Carlo tree search-guided optimization.