model-agnostic framework
A model-agnostic framework enables the development and application of algorithms that are not tailored to specific models, allowing broader applicability across different architectures and learning paradigms.
- APOLLO: Automated LLM and Lean Collaboration for Advanced Formal Reasoning
- Cue3D: Quantifying the Role of Image Cues in Single-Image 3D Generation
- DiCoFlex: Model-Agnostic Diverse Counterfactuals with Flexible Control
- Distributionally Robust Feature Selection
- Enforcing Hard Linear Constraints in Deep Learning Models with Decision Rules
- Improving Time Series Forecasting via Instance-aware Post-hoc Revision
- KAIROS: Scalable Model-Agnostic Data Valuation
- Knowledge Distillation Detection for Open-weights Models
- MAGNET: A Multi-agent Framework for Finding Audio-Visual Needles by Reasoning over Multi-Video Haystacks
- Modality-Aware SAM: Sharpness-Aware-Minimization Driven Gradient Modulation for Harmonized Multimodal Learning
- OmniFC: Rethinking Federated Clustering via Lossless and Secure Distance Reconstruction
- ScaleDiff: Higher-Resolution Image Synthesis via Efficient and Model-Agnostic Diffusion