Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
FOCUS: Training-Free Decision-Preserving Context Compression for LLM AgentsMicrosoft · arXiv (Cornell University)Sep 29VADER: Filtered Vector Search with Declarative RecallAlphabet · arXiv (Cornell University)Sep 29HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model DeploymentLG AI Research · arXiv (Cornell University)Sep 29VideoLoop: Looped Working Memory Against Semantic Thrashing in Long-Form Video AgentsMicrosoft · arXiv (Cornell University)Sep 29Designing Reliable LLM-as-a-judge Measurement Systems for Multi-turn Business AgentsMeta · Research SquareSep 29Measuring Service Congestion, Sharing, and Pacing in a Residential Access NetworkAlphabet · ACM Internet Measurement Conference (IMC)Sep 29Towards Vision‐Language Geo‐Foundation Models: A SurveySenseTime · The Photogrammetric Record3 citationsSep 28On nonlinear dimensionality reduction, linear smoothing, and boundary conditionsMeta · Science China Information SciencesSep 28What Visual Generators Need from Teachers: Rethinking Representation AlignmentSenseTime · arXiv (Cornell University)Sep 28Language Models Are "Insecure" ReportersAlphabet · arXiv (Cornell University)Sep 28Dyad: Extending Large Language Models with Native Typed Decision-MakingMicrosoft · arXiv (Cornell University)Sep 28Targeting Pivotal Decisions for Credit Assignment in Agentic Reinforcement LearningMicrosoft · arXiv (Cornell University)Sep 28HeurEvo: Agentic Evolution of Hybrid Solver-Augmented Heuristics for Time-Critical Mathematical OptimizationMicrosoft · arXiv (Cornell University)Sep 28The Signed Geometry of One-Shot Recourse: On-Path Validity and the Signed-Curvature CriterionAlphabet · arXiv (Cornell University)Sep 28The Universal Classifier for Graph LearningAlphabet · arXiv (Cornell University)Sep 28Automated Pre-Silicon Verification of High-Speed DDR5 and LPDDR5/6 Memory Controllers: Closed-Loop Timing, Mode Register, and PHY Synchronization in UVMAlphabet · arXiv (Cornell University)Sep 28Wavel: A Fast and Efficient Compilation System for Wafer-Scale AcceleratorsMicrosoft · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)Sep 28UniOPSD: Unifying Outcome and Hindsight Feedback for Agentic Reinforcement LearningMininglamp Technology · arXiv (Cornell University)Sep 28Evolving Support Priorities in Empathetic Reinforcement LearningSenseTime · arXiv (Cornell University)Sep 28StreamEP: Straggler-Tolerant MoE Decoding without Communication BarriersAlphabet · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)Sep 28Long Time No See: Benchmarking VLMs for Out-of-Sight Spatiotemporal Reasoning in Egocentric VideosMicrosoft · arXiv (Cornell University)Sep 28A Few GPUs, A Whole Lotta Scale: Faithful LLM Training Emulation with CrystalLLMAlibaba · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)Sep 28Cerium: A Multi-GPU Framework for Terabyte-Scale Encrypted InferenceNVIDIA · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)Sep 28Beyond Verbalized Confidence: Calibrating Reasoners with Differentiable ReadoutsAlibaba · arXiv (Cornell University)Sep 28Teacher-Student Gaps Are Not Enough: Outcome-Guided On-Policy Distillation for Multi-Turn Autonomous AgentsAlibaba · arXiv (Cornell University)Sep 28