Archive
Everything we've scored that's past the 48-hour window — open to everyone, no account needed. The live feed only shows what's moving right now; this is the record behind it.
- 19BS: Take the Hint - Interactive Multitracer PET/CT Lesion Segmentation with a Scribble-Conditioned ResEnc U-Net
- 19A Model with No Head and Many Thoughts
- 19Retrieved but not ranked: surface-form bias in structural retrieval, from mathematics to agent trajectories
- 19Distinct dynamics of conceptual and referential disruptions in human reading and large language model processing
- 19Agentic Harnesses: LLM-Driven Verification Layers for Robot Autonomy
- 19Gradient-Update Mismatch: Rethinking Conflict-Free Training of Physics-Informed Neural Networks
- 19Evaluating Verified Autonomy in Quantum Engineering
- 19A Cascaded Unsupervised-Supervised NLP Pipeline for Detecting Accusatory Language in Public Procurement
- 19CardioFusion-AI: Robust ECG--PPG Fusion for Multimodal Physiological Monitoring Under Signal Degradation
- 19Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models
- 19The Interaction Tax: When Communication Erases Diversity in Multi-Agent Teams
- 19Enhancing EBSD throughput of battery electrode materials using super-resolution generative adversarial networks
- 19Real-Time Video Anomaly Detection Using YOLO Pose Estimation and CLIP-Based Semantic Scoring
- 19H3-World: Turning Language Understanding into World Control
- 19Earth observation embeddings are effective sub-grid descriptors for probabilistic weather downscaling
- 19How AI Assistance Affects Human Skill Development: A Study of Learning with Logic Puzzles
- 19AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMs
- 19Multi-Task Learning for Sparsely-Labeled Time Series: A Case Study on Cold-Hardiness Modeling
- 19VoiceMem: Streaming Dual-Brain Memory for Real-Time Interaction
- 19Shopping by algorithm: How agentic AI deploys human heuristics as a surrogate consumer
- 19Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence
- 19Continuous-Time Reinforcement Learning for Controlled Hawkes Jump-Diffusions
- 19Emergent Collusion in Long-Horizon LLM Agent Interaction
- 19Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents
- 19Test-Time Self-Evolving GUI Visual Grounding via Reflection-Guided On-Policy Self-Distillation
- 19Rare Event Estimation via Iterative Unalignment
- 19Towards Expert-level Medical AI for Real-time Video Consultations
- 19Optimal Low-Rank Quantum State Tomography with Bounded-Sample Joint Measurements
- 19Convergent Detour Hijacking: Task-Preserving Resource Amplification in Skill-Based LLM Agents
- 19A Self-Evolving Multi-Agent Framework Defense against LLM Jailbreak Attacks
- 19DolphinBench: Mapping the Pareto Frontier of Agent Memory
- 19RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
- 19A Neighborhood Attention Transformer Network for Enhanced 3D Segmentation of the Left Anterior Descending Artery
- 19Harness-Zero: Harness Distillation via Agent-as-Harness
- 19Inertial Manifold Neural Operator for Dissipative Time-Dependent Partial Differential Equations
- 19GEO-Flag: Detecting and Measuring GEO-Optimized Web Content
- 19DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation
- 19Robustness of Anomaly Detection Models for Industrial Control Systems under Training-Time Data Contamination
- 19How Much is a Human Right Worth? ECtHR-NPD: A Benchmark for Predicting Non-Pecuniary Damage Awards
- 19Beyond Outcomes: Dual-View Relational Learning for Efficient Agent Benchmarking
- 19Imitation Learning for Connection-Tableau Construction
- 19Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication
- 19StagedWorkspace: A Versioned Workspace for Knowledge-Work Agents
- 19Reconciling Process Supervision with Outcome-Based Credit in Agentic Policy Optimization
- 19LoRA-generating hypernetworks for efficient on-device LLM generative personalization
- 19Interpretable AI predicts a 2026 summer dry anomaly in central China
- 19Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration
- 19Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization
- 19onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction
- 19WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory
- 19ChildSafeAds Shared Task 2026: Commercial Content in Child-Facing YouTube Videos
- 19Critical-State RL: Diagnosing Trainable States for Multi-Turn Tool Use
- 19Token-Efficient Data Reasoning Agents via Adaptive Structuring of Unstructured Data
- 19CaliBench: Are the Stochastic Dynamics of Video World Models Physically Calibrated?
- 19VISA: Agentic Self-Evolving Data Synthesis for Multimodal Instruction Following
- 19Structural Silence: When AI Infrastructure Fails Speakers of Underrepresented Languages
- 19Learned, Then Lost: A Measured Single-Example Counterfactual in Pre-training
- 19Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders
- 19Show-Harness: Just a VLM Agent Can Play Robots
- 19How My AI Agents' Mistakes Become Permanent Rules (And Why I Want Them to Fail)
- 19Stealing Reasoning Traces from Proprietary LLM APIs
- 19ForgetMimic: Motion Unlearning for Reinforcement Learning Humanoid Control
- 19Characterizing Language Generation in the Limit: Finite Witnesses and a Separation-Width Hierarch
- 19GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay
- 19Logarithmic-Free Moment and Generalization Bounds for Uniformly Stable Algorithms
- 19Primitive Representation Learning for Unsupervised Dynamic Contrast Enhanced MRI Reconstruction
- 19Policy Iteration with Human Feedback: Bringing Post-Training RL to In-context Learning
- 19ConvergeFlow: Language Flow with Provable Convergence to Token Embeddings
- 19ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls
- 19Prime Agent: A Self-Improving RLM Harness
- 19The First Token Is a Clue: Verbalizing Multi-Token Concepts from the J-lens
- 19A systematic Approach to constructing a Chance-and-Risk Matrix for Semiconductor Supply Chains
- 19Performance of Clinical AI System and Physicians and Frontier Language Models in primary care diagnostics
- 19Provably adaptive sampling with uniform and remasking discrete diffusion models
- 19HLSR: Hybrid Live Forecast Selective Dynamic Vehicle Rerouting for Real-Time Congestion Avoidance
- 19From Confusion to Clarity: Confusion-Aware Retrieval and Knowledge Injection for Text Classification
- 19Time-Aware Validation of Machine Learning Fuel Consumption Models: Evidence from 1\,Hz Operational Data, CCGS \textit{Sir Wilfrid Laurier}
- 19ToolLoop: Closed-Loop Tool-Use Data Synthesis via Decomposed Generation and Dynamic Self-Feedback
- 19Higher-order pruning of experts in mixture-of-experts language models
- 19Diffusion-Induced Spatial Attention Overlapping Community Detection
- 19Sci-VBench: Evaluating Knowledge- and Reasoning-Intensive Video Generation in Science Domains
- 19Surgical WAM: A World-Action Model for Data-Efficient Surgical Robot Learning
- 19Rethinking Circuit Evaluation: Do Circuits Explain Model Errors?
- 19A positive resolution of the gap-entropy conjecture
- 19ArchAgent v2: A Case Study with the Data Prefetching Championship
- 19Lévy Attention: Single-Pass Predictive Uncertainty for Continuous-Time Attention
- 19Continual Concept Personalization in Diffusion Models via Residual-Space Adaptation
- 19Delegation Asymmetry in Agentic Recommender Systems: Measuring Two-Sided Receptivity in Online Dating
- 19When and Where to Trust the Teacher: Unifying On-Policy Distillation and GRPO through Entropy-Calibrated Credit Assignment
- 19Likelihood-free inference with nuisance parameters through normalizing flows
- 19Model Hypnosis: Strong control of AI via additive subliminal effects
- 19Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers
- 19DualOPSD: Adaptive Privileged Teachers for On-Policy Self-Distillation
- 19The concentration game: Bayesian updating, regret, and information
- 19Minimax bounds for watermarked and masked recursive discrete distribution estimation
- 19IdeaAMBIG: Benchmarking Implementation-Critical Gaps in Research-Idea Specifications
- 19Finetuning Strategies for Querying Sounds by Vocal Imitation
- 19Evolving Semantic Indexes for Generative Retrieval
- 19TokEval: A Tokenizer Evaluation Suite
- 19Physics-Constrained Deep Learning Model for Contactless Blood Pressure Monitoring from Triaxial Bodyseismography
- 19EG-ARSA: An Expert-Grounded Open Model for Visual Road Safety Auditing in Low-Resource Settings
- 19SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?
- 19ReWorld: An Interactive World Model with Long-Horizon Memory
- 19Distribution Matching Distillation for Continuous Diffusion Language Models
- 19ThinkPrior: Zero-Rollout Difficulty Priors for Cold-Start Prompt Selection in RLVR
- 19ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation
- 19Comparison of techniques for fine-tuning open-weight models for entity extraction from radiology reports
- 19Energy-Structured Latent World Models with Neural Time Fields for Physically Constistent Open-World Motion Planning
- 19How to Train a Critic Stably and Efficiently
- 19HAF: Adapting Generalist VLAs to Humanoid Whole-Body Loco-manipulation via Hierarchical Action Flow and Spectral Latent RL
- 19Report: Progressive Disclosure of Agent Skills
- 19Reading Is Not Using: Retrieval, Judgment, and the Design of AI Financial Research Workflows
- 19Reasoning with Continuous Latent Diffusion
- 19Why does getObject() throw ORA-17004 for a VECTOR OUT parameter?
- 19Rethinking Personalized Generation: Test-Time Alignment via Factorized Ranking Models
- 19I Found a Hidden Layer Inside AI Images
- 19Quantum score matching with applications to learning thermal states
- 19Everything in Moderation: Per-Domain Coverage Optima and Alignment-Resistant Domain Gaps in Multi-Domain Mid-Training
- 19On the Fragility of Self-Improving Agents: Variance, Task Order, and Underspecification
- 19GoDeep: Annotation-Free Open-Vocabulary 3D Scene Understanding via Language-Space Lifting
- 19Automatic depth-based local center clustering via $β$-integrated local depth and adaptive grouping
- 19Beyond Teacher Likelihood: Group-Calibrated On-Policy Distillation for Long-Context Reasoning
- 19Financial Numerical Prediction and Allocation as Token Generation
- 19Provable Benefits of Regularization: Fast Rates for Adversarial Imitation Learning
- 19ADEPT: Accelerating Dexterity via Pre-Training and Post-Training using Reinforcement Learning
- 19It's Not RoPE that Creates Sinks: The Role of Self-Concentration and Value-Non-Mixing in Attention
- 19SP-HLM: Sparse-Pose-Driven Motion Generation and Whole-Body Control for Humanoid Loco-Manipulation
- 19Fine-Tuning LLMs for Translation: General Forgetting Mitigation Does Not Preserve MT-Specific Instruction Following
- 19Metrics Failure in LLM-Based Code Vulnerability Repair: An Empirical Study and a Change-Aware Screen
- 19Distillation Defenses Easily Break After Reinforcement Learning
- 19One Adapter, Many Tasks: Task-Conditioned Feature Transformations for Continual Learning
- 19Which Differences Matter? Controlled Tests of Rule Transfer, Joint Reliability, and Information Use in Decision Models
- 19STARS: From Spatiotemporal Dynamics to Social Representations in Human-Robot Interaction
- 19Cross-Regional Grapevine Cold Hardiness Prediction via Learned Multimodal Latent Representations
- 19Decomposition Buys Integrity, Not Yield
- 19VAKRA: Evaluating Multi-Hop Reasoning Across APIs and Retrieval Under Tool-Use Policies
- 19The Token-Energy Law of LLM Decode: Active Bytes and Realized Time Account for Joules per Token
- 19Multi-Agent AI System for Radiology Report Structuring and Quality Assurance with Independent Radiologist Evaluation
- 19EquivSVA: A Formally Verified Dataset of Behavioral Assertions Across Equivalent RTL Implementations
- 19LAION-BVD: A 10-Million-Hour Open Video Dataset for Multimodal Pre-training
- 19MeqMuon: Matrix-Equilibrating Muon for LLM Pretraining
- 19Cache Engineering for Agents
- 19FedV-KGQA: Multi-Hop Question Answering over Vertically Partitioned Knowledge Graphs
- 19Large Language Model-Driven Small-Capitalization Trading: Integrating Financial News Sentiment, Macroeconomic Indicators, and Technical Signals
- 19Memory Attention
- 19Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
- 19SHE: Trajectory-driven Safety Harness Evolution for LLM Agents
- 19Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics
- 19Comprehensive reconstruction of collider events with hypergraph representation learning and graph-conditioned diffusion
- 19From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix
- 19Continuous-Time Policy Gradient from Discrete Data: Convergence Analysis and Model-Free Algorithms
- 19S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?
- 19Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs
- 19Measuring LLM Sycophancy under Sustained Multi-Turn Pressure
- 19A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion
- 19BrowserForge: Scaling Web Episode via Parallel Browser Sandboxes
- 19From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation
- 19Closing Cost-Quality Gap in Document VLMs: Difficulty-Aware Data Curation and Quality-Adjusted Deployment Economics
- 19SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User Simulation
- 19Social Laws for Multi-agent Coordination in Stochastic Environments
- 19Reinforcing Agentic Creativity in Scientific Ideation with Night Science
- 19SPADE: Self-Play in Adaptive Synthetic Executable Environments
- 19Form-Associated Custom Elements: Web Components That Belong in a Form
- 19ScAn-Bench: Evaluating Scaling Analysis Methodology
- 19The Surprising Effectiveness of Approximate Value Iteration in Self-Play
- 19LLM Post-Training as Brownfield Maintenance: An Industrial Perspective on Dataware Engineering
- 19ComputerSD: Online Self-Distillation from Real-Time Feedback for Computer-Use Agents
- 19Space-Creating versus Dead Possession: An Off-Ball Possession-Quality Index for Broadcast Football
- 19BDH-CQ: In-Context Learning with Recurrent Latent Reasoning
- 19FleXray: Universal Clinical X-ray Segmentation
- 19Learning Collective Dynamics with Differentiable Gaussian Representations
- 19Proteus: Incremental Memory Activation for Long-Context Sequence Modeling
- 19Type-Safe Is Not Error-Free: A Constrained Decision Head Follows the Option Name, Not the Rubric Bound to It
- 19X-Reset: Scaling Object-Centric Reinforcement Learning via Cross-Embodiment Resets
- 19Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics
- 19Long-Lived Characters, Local Inference: Incremental Memory Maintenance for Game NPCs
- 19BLOOM-WILT: Logit Tilting for Behaviour Elicitation in Automated LLM Auditing
- 19Where Do LLM Ensemble Gains Come From? Decomposing Candidate Value, Diversity, and Aggregation
- 19Bytes to Threats: Byte-Sequence Transformer For Malware Detection
- 19Fusion Training for Mathematical Generalization in Large Language Models
- 19Learning Holographic Reduced Representations with Clifford Variational Autoencoders
- 19GeneICL: A Tabular Foundation Model for Bulk Transcriptomics
- 19From Reasoning to Action: Benchmarking Frontier AI Models on LIBERO-10 Robot Manipulation Tasks
- 19Beyond the Sequentiality Bottleneck: A Hybrid Diffusion-Autoregressive Framework for Efficient Multimodal Reasoning
- 19Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence
- 19SceneRig: An Agentic System for Simulation-Ready 3D Scene Reconstruction from Single Images
- 19Beyond Local Surprise: Grounded Dialogue as Selective Belief Revision under Referential Uncertainty
- 19Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents
- 19OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning
- 19Capable Teacher, Myopic Scorer: Reward-Free Teacher Repair for Agentic On-Policy Distillation
- 19Two $\operatorname{ReLU}$ Layers is All You Need
- 19A2M: Trace-Optimized Agent Hijacking in the MCP Ecosystem
- 19Stress-Testing Efficient Responsible-AI Evaluation: When Compute Savings Change Benchmark Conclusions
- 19Belief-Aware Multi-Agent Path Finding under Map Uncertainty
- 19Don't Be Nice
- 19Probe-Space Preconditioning for Fast and Stable Zero-Order Training
- 19Trace Integrity for LLM Data Agents: A Vision for Auditable Structured Reasoning in Real-World Systems
- 19Tail-Influence Sampling for CVaR Policy Evaluation
- 19WildMAE: Structure-Aware Masked Autoencoding for Wildlife Image Reconstruction
- 19NeuronEye: Query-Guided Visual Concept Activation for Vision-Language Reasoning