Rocket - Think it. Type it. Launch it.
Mini toolsPapers
Filter by company
-
Moloch's Bargain: Emergent Misalignment When LLMs Compete for AudiencesStanford University
-
SuffixDecoding: Extreme Speculative Decoding for Emerging AI Applications
-
Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy (short paper)The Pennsylvania State University
-
Less is More: Recursive Reasoning with Tiny Networks
-
Tongyi DeepResearch Technical Report
-
TabArena: A Living Benchmark for Machine Learning on Tabular Data
-
IBM Granite 4.0: hyper-efficient, high performance hybrid models for enterprise
-
Diffusion Adversarial Post-Training for One-Step Video Generatio
-
Pretraining Large Language Models with NVFP4
-
Training Agents Inside of Scalable World Models
-
APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-tail Generation
-
WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
-
Inference-Time Scaling for Generalist Reward Modeling
-
MechStyle: Augmenting Generative AI with Mechanical Simulation to Create Stylized and Structurally Viable 3D ModelsGoogle Research, Stability AI / Center for Bits and Atoms, MIT, Khoury College of Computer Sciences, Northeastern University, University of Washington
-
FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
-
Dynamic Parameter Memory: Temporary LoRA-Enhanced LLM for Long-Sequence Emotion Recognition in Conversation
-
ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning
-
From Slow Bidirectional to Fast Autoregressive Video Diffusion Models
-
EpiCache: Episodic KV Cache Management for Long Conversational Question Answering
-
"My Boyfriend is AI": A Computational Analysis of Human-AI Companionship in Reddit's AI CommunityMassachusetts Institute of Technology
-
Teaching LLMs to Plan: Logical Chain-of-Thought Instruction Tuning for Symbolic Planning
-
Towards an AI-Augmented Textbook
-
Steering MoE LLMs via Expert (De)Activation
-
Robix: A Unified Model for Robot Interaction, Reasoning and Planning
-
AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data
-
An AI System to Help Scientists Write Expert-Level Empirical Software
-
Why Language Models Hallucinate
-
Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving
-
GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
-
Measuring the environmental impact of delivering AI at Google Scale
-
3D-GENERALIST: Vision-Language-Action Models for Crafting 3D Worlds
-
X-MoE: Enabling Scalable Training for Emerging Mixture-of-Experts Architectures on HPC Platforms
-
NoProp: Training Neural Networks without Full Back-propagation or Full Forward-propagation
-
Matrix-3D: Omnidirectional Explorable 3D World Generation
-
Amazon Ads Multi-Touch Attribution
-
Scaling Laws for Native Multimodal Models
-
Devstral: Fine-tuning Language Models for Coding Agent Applications
-
Establishing Best Practices for Building Rigorous Agentic Benchmarks
-
No LLM Solved Yu Tsumura's 554th ProblemUniversity of Cambridge, University of Oxford
-
Why do LLMs attend to the first token
-
Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction
-
Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation
-
Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference
-
Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems
-
Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks
-
Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving
-
Kimi K2: Open Agentic Intelligence
-
Scaling Data-Constrained Language Models
-
Seed LiveInterpret 2.0: End-to-end Simultaneous Speech-to-speech Translation with Your Voice
-
Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them
