Papers
Videos
Whiteboards
Open Problems
Email Digest
Labs
Pricing
Updates
Log in
Sign up
Research Explorer
Papers
Videos
Whiteboards
Open Problems
Email Digest
Labs
Pixel Art Generator
All Projects
Pricing
Log in
Sign up
Discord
Discord Logo Streamline Icon: https://streamlinehq.com
Whiteboard Explanations
Self-Preservation Bias in LLMs
Simple Self-Distillation for Code Generation
ByteRover: Agent-Native Memory for LLMs
Multimodal Analysis of Israel-Hamas War on YouTube Shorts
Screening Is Enough: Multiscreen Model
Nuclear Test-Transient Link in POSS-I Plates
Transformers and Attention for Applied Mathematics
Self-Organizing LLM Agents Outperform Structures
LLM Withdrawal: Impact on Knowledge Workers
Short Proofs in Combinatorics & Number Theory
TurboAngle: KV Cache Compression via Angle Quantization
HandX: Scalable Bimanual Hand Motion Generation
Shor's Algorithm with 10,000 Atomic Qubits
Meta-Harness: End-to-End Harness Optimization
Fragility in LLM Social Networks
Frequency-Time Diffusion with Neural Cellular Automata
Mathematics and AI: Rethinking Human Thought
Quantifying Frontier LLM Capabilities for Container Sandbox Escape
ARC-AGI-3: New AGI Benchmark
MARCUS: Agentic Multimodal Cardiac Diagnosis
Kitchen Loop: Autonomous Code Evolution
Hamilton Decompositions of the Directed 3-Torus
Voxtral TTS: Hybrid AR & Flow-Matching TTS
Evaluating LLMs for Harmful Manipulation
SlopCodeBench: Degradation in Iterative Coding Agents
EVA: Efficient RL for Video Agents
MegaFlow: Zero-Shot Large Displacement Flow
Natural-Language Agent Harnesses
Confidence Mesh Extraction from 3D Gaussians
Sign Errors in Black Hole Mechanics
MIRAGE: The Illusion of Visual Understanding
SpectralSplats: Robust Differentiable 3D Tracking
Claudini: Autonomous Adversarial LLM Attacks
Self-Distillation Effects on LLM Math Reasoning
Silent Memory Pollution in Claw AI Systems
PERMA: Benchmarking Personalized Memory Agents
RealMaster: Lifting Rendered Scenes to Photoreal Video
Efficient Policy Optimization via Delight-Driven Gating
Finetuning Activates Verbatim Recall in LLMs
PRISM: O(1) Photonic KV Cache Selection in LLMs
UNITE: Unified Tokenization & Latent Denoising
PivotRL: Efficient Agentic LLM Post-Training
AI Data Agents and DAB Benchmark
Fast Astronomical Transients in Archival Plates
CLT-Forge: Scalable Cross-Layer Transcoder Library
LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from...
Autonomous AI in Experimental HEP Analysis
Hyperagents: Recursive Metacognitive Self-Improvement
Distilling Discrete Diffusion Models with D-MMD
Lambda-RLM: Y-Combinator for LLMs
Golden-Ratio Info Balance & Antifragility
PRISM: Efficient Persona Routing in LLMs
Secure Linear Alignment of LLMs
Chronos: Temporal-Aware Conversational Memory
MOND Depth Index & Maturity Clock
Transformers and Cortical Microcircuits
SmartSearch: Ranking Beats Structure for Memory Retrieval
Schrödinger Bridges for Generative Modeling
R-Equivalence on Cubic Surfaces: Non-Trivial Cases
Nemotron-Cascade 2: Advanced LLM Post-Training
LLMs and the Distortion of Written Language
Memento-Skills: Autonomous Skill Learning
Transformers as Bayesian Networks
MosaicMem: Hybrid Memory in Video Models
Local LLM Agents for Linux Privilege Escalation
MetaClaw: Continual Meta-Learning for LLM Agents
Delightful Policy Gradient: Noise Suppression in RL
Delusional Spirals in LLM Dialogues
Ghosts of Softmax: Singularities in Cross-Entropy
NanoGS: Training-Free Gaussian Splat Simplification
Fast-WAM: Rethinking Test-Time Imagination
GhanaNLP: Multilingual Resources for Ghanaian Languages
Autonomous AI: Cognitive Science Roadmap
PokeAgent Challenge: Competitive & Long-Context AI
HorizonMath: AI Mathematical Discovery Benchmark
V-JEPA 2.1: Dense Features in Video SSL
Emergent Artifact Exchange in Autonomous Agents
Mamba-3: Advanced State Space Sequence Models
Attention Residuals in Deep Language Models
Mixture-of-Depths Attention in LLMs
Energy-Based Fine-Tuning for Language Models
Agent-to-Sim: Learning Interactive Behavior Models from Casual Longitudinal...
AI Progress in Multi-Step Cyber Attacks
Evaluation Format Drives Triage Failure in Health AI
GLM-OCR: Compact Multimodal OCR Model
Agentic Reasoning in Multimodal Documents
AutoResearch-RL: Autonomous Neural Architecture Discovery
DVD: Deterministic Video Depth Estimation
Simple Recipe: VLA Models as Continual Learners
Neural Thickets in Pretrained Models
LM Head as a Gradient Bottleneck
OpenClaw-RL: Online RL for Personalized Agents
RL-Augmented Teleoperation for Dexterous Manipulation
TiPToP: Modular Open-Vocabulary Robotic Manipulation
AgentOS: NL-Driven OS Paradigm
PostTrainBench: Evaluating Automated LLM Post-Training
Covenant-72B: Decentralized LLM Pre-Training
Reinforced Generation of Ramsey Numbers
3I/ATLAS: Isotopic Evidence for Cold, Ancient Formation
AMIE: Conversational Diagnostic AI in Primary Care
Show More
