Guy

The paper introduces Clover, a code completion tool that logs students' interactions and offers attention checks to promote reflective engagement during programming tasks.

Researchers used a large language model to generate checkpoint/restart code for MPI applications, achieving comparable efficiency to human-engineered solutions.

This paper introduces the Bridge, a lightweight interface that enables distillation between Teachers and Students with different latent resolutions and VAE spaces in Cross-Space Distillation.

Kani is an open-source model checker for Rust that provides correctness guarantees for safety properties through compilation and a specification language.

This paper presents a memory-efficient training stack for Mixture-of-Experts (MoE) models, combining and specializing parallelism techniques for maximal efficiency.

This paper introduces BamiBERT, a new Vietnamese language model based on BERT that addresses limitations of PhoBERT and sets a new state-of-the-art among base-sized Vietnamese encoders.

This paper proposes a Real-Bogus classification framework using injection-driven, weakly supervised training without human-labeled data, achieving strong performance and calibrated uncertainties.

The paper introduces BACH, a multi-interest two-tower retrieval model that uses a per-user mixture over heads, mitigating collapse and producing a per-user weighting of interests.

A multimodal Mixture-of-Experts framework was developed for asthma detection using vocal biomarkers and clinical data, achieving better performance than unimodal and bimodal approaches.

This paper presents a sentence-level sign language translation system with real-time deployment, using a fine-tuned SHuBERT-ByT5 translation stack and a hardware-aware streaming system.

This paper presents TerraRepair, a tool-grounded LLM agent for Terraform repair with structured escalation, which improves scanner-verified fix rates for IaC security scanners Checkov and Trivy on AWS, Azure, and GCP.

This paper proposes an end-to-end AI-accelerated framework for upskilling workers, validated by industry partnerships and exam success.

A self-evolving annotation framework for Major Depressive Disorder using large language models and expert verification is proposed to improve annotation consistency and explainability.

This paper presents the formal semantics for unfolding expressions and pure functions in deductive program verifiers for increased modularity.

CLIFE is an edge-native camera-LiDAR fusion framework that enhances perception of vulnerable road users under varied environmental and traffic conditions, with adaptive calibration, O(N log N) per-frame cost, and high throughput.

This paper proposes a Bayesian Optimization-based framework for multipath QUIC (MPQUIC) scheduling that jointly considers maximum upload completion time and total LTE usage, identifying Pareto-efficient operating points.

This paper introduces GraphVid, a graph-conditioned image-to-video generation model enabling precise multi-subject control through structured interaction graphs, and curates GraphVid-Bench, a large-scale interaction-centric video dataset.

This paper introduces online assignment policies for stochastic matching on hypergraphs, which are maximally stable and allow for the derivation of necessary and sufficient stability criteria.

This paper proposes QCOEM, a quantum cloud orchestration framework using evolutionary algorithms for multi-objective optimization of quantum task scheduling, achieving zero rescheduling and 30% higher mean fidelity than noise-agnostic heuristics.

Researchers conducted shadow evaluations on AI agents' ability to carry out open-ended AI research by having them work on unpublished NeurIPS papers and having the original authors grade their output. Agents struggled with critical parts of the research lifecycle.