Papers
Videos
Whiteboards
Open Problems
Email Digest
Pricing
Log in
Sign up
Discord
Discord Logo Streamline Icon: https://streamlinehq.com
Updates
Papers
Videos
Whiteboards
Open Problems
Email Digest
Pricing
Log in
Sign up
Discord
Discord Logo Streamline Icon: https://streamlinehq.com
Whiteboard Explanations
Pass@k Optimization and Pass@1 Degradation in LLMs
SymTorch: Symbolic Distillation for Deep Learning
VGG-T3: Scalable Offline 3D Reconstruction
PhysicEdit: Physics-Aware Image Editing
Expert Investment Teams via Multi-Agent LLMs
ISO-Bench: Optimizing Inference Workloads
DualPath: Breaking Storage Bottleneck in LLM Inference
Agent-Based Platforms: OpenClaw, Moltbook, ClawdLab
RL Training for Deep Research Agents
Economics of AGI: The Measurability Gap
Transformers Learn Transfer Operators In-Context
Turing Completeness of GNU find
Aletheia Autonomous Proofs on FirstProof
TOPReward: Token-Based Zero-Shot Robotic Rewards
Agents of Chaos: LLM Agent Failures
Agentic Alpha: Autonomous Market Signals
Learning Without Training: Paradigm Shift
Geometry of Noise: Autonomous Diffusion
LLM Decoding via Probability Simplex Optimization
Human-Centric XR: Hand & Camera Control
Adaptive Test-Time Compute Allocation
FAMOSE: ReAct-Driven Auto Feature Discovery
Large Underground Xenon (LUX) Experiment
MoltNet: Social Behavior of AI Agents on MoltBook
B3-Seg: Camera-Free 3DGS Segmentation
Online Deanonymization via LLMs
Reverso: Efficient Models for Zero-shot TS Forecasting
OpenClaw Agents: Risky Sharing & Norm Enforcement
Unified Latents: Training Latent Diffusion Models
Fast KV Compaction via Attention Matching
LLM Assistance in Novice Biology Labs
Softmax Information Geometry: Probing & Steering
Perceptive Humanoid Parkour: Dynamic Skill Chaining
In-Context Co-Player Inference in Multi-Agent RL
Calibrate-Then-Act: Cost-Aware LLM Exploration
TraceRouter: Robust Safety for Large Foundation Models via Path-Level...
Reliable and Responsible Foundation Models: A Comprehensive Survey
FactorMiner: Self-Evolving Agent for Alpha Discovery
Rise of AI Agent Communities on Moltbook
Masked Updates in Adaptive Optimizers
GLM-5: From Vibe Coding to Agentic Engineering
Symmetry in Language: Geometric Embeddings
AnchorWeave: World-Consistent Video Generation
Experiential Reinforcement Learning
Socialization in AI Agent Societies
Sphere Encoder for Fast Image Generation
Boule vs Baguette: Task Topology & Reasoning
Hidden Risks in American Options
BitDance: Scaling AR Models with Binary Tokens
BEACONS: Neural PDE Solvers with Formal Guarantees
Soft Contamination in LLM Benchmarks
Evaluating AGENTS.md Impact on Coding Agents
SkillsBench: Evaluating Agent Skills
Data Repetition Beats Data Scaling in Long-CoT SFT
CoPE-VideoLM: Efficient Video Language Models
SWE Context Bench: Context Learning Benchmark
AI Coding Agents: PR Acceptance Analysis
Integrating Code Metrics for Notebook Docs
Rethinking Code Complexity for LLMs
SWE-AGI: Spec-Driven Software Construction
Causal-JEPA: Object-Level Latent Interventions
Pedagogical Synthesis for Language Model Distillation
Collective AI Behavior on Moltbook
Categorical Flow Maps: Accelerated Discrete Generation
Retrieval-Aware Distillation for Transformer Hybrids
Nonzero Single-minus Gluon Tree Amplitudes
Spin Splitting and Kondo in InSb QD Nanosheets
Gossip-Driven Indirect Reciprocity in LLM Agents
Intelligent AI Delegation Framework
Self-Referential Vocabulary-Activation in LLMs
Olmix: Data Mixing for LM Development
Authenticated Workflows Protecting Agentic AI
Pinnacle Architecture: QLDPC for RSA-2048
SoftMatcha 2: Trillion-Scale Pattern Matching
YOR: Affordable Mobile Manipulator
Autonomous Mathematics Research
Geometric Flow in Diffusion Transformers
QuantaAlpha: Evolutionary LLM Alpha Mining
Detecting Unverbalized Biases in LLMs
Neural Scaling Laws in Language Models
EvoCorps: Evolutionary Online Discourse Governance
SkillRL: Recursive Skill-Augmented RL
OneVision-Encoder: Codec-Aligned Multimodal Intelligence
Hybrid Gated Flow for 1.58-bit LLMs
Biological Blueprint for Machine Intelligence
LLMs and Code Security Challenges
LLM Reasoning Failures Survey
AIRS-Bench: Evaluating Autonomous Research Agents
Generative Meta-Models for LLM Activations
From Kepler to Newton: World Models in Transformers
Learning Rate in LoRA Fine-tuning
LLM Multi-Agent Teams Underperform Experts
Shared LoRA Subspaces for Continual Learning
Structured Context Engineering for File-Native Systems
LLM Agents for Root Cause Analysis
Authorship Drift in LLM Writing: Efficacy & Trust
Privileged Info Distillation for LMs
Computing Diffusion Geometry
First Proof: AI Evaluation in Research Math
ERNIE 5.0: Trillion-Parameter Multimodal Model
Show More
