Few-shot Steerable Alignment: Adapting Rewards and LLM Policies with Neural Processesby Katarzyna Kobalczyk, Claudio Fanconi,…
AnchorInv: Few-Shot Class-Incremental Learning of Physiological Signals via Representation Space Guided Inversionby Chenqi Li, Boyan…
Extreme Multi-label Completion for Semantic Document Labelling with Taxonomy-Aware Parallel Learningby Julien Audiffren, Christophe Broillet,…
Efficient Fine-Tuning of Single-Cell Foundation Models Enables Zero-Shot Molecular Perturbation Predictionby Sepideh Maleki, Jan-Christian Huetter,…
Fine-tuning Aligned Classifiers for Merging Outputs: Towards a Superior Evaluation Protocol in Model Mergingby Fanshuang…
An Agentic Approach to Automatic Creation of P&ID Diagrams from Natural Language Descriptionsby Shreeyash Gowaikar,…
SceneDiffuser: Efficient and Controllable Driving Simulation Initialization and Rolloutby Chiyu Max Jiang, Yijing Bai, Andre…
Meta-Controller: Few-Shot Imitation of Unseen Embodiments and Tasks in Continuous Controlby Seongwoong Cho, Donggyun Kim,…
Semi-Supervised Risk Control via Prediction-Powered Inferenceby Bat-Sheva Einbinder, Liran Ringel, Yaniv RomanoFirst submitted to arxiv…
On Distilling the Displacement Knowledge for Few-Shot Class-Incremental Learningby Pengfei Fang, Yongchun Qin, Hui XueFirst…