Skip to content
conftrace_

performance gain

6 papers

Explore in graph

Co-occurring keywords · papers mentioning both

topological augmentation (1) domain-adaptive routing (1) domain-expert synergy loss (1) mixture-of-domain low-rank experts (1) misfit (1) model kinship (1) continual merging (1) benchmark performance (1) plug-and-play module (1) continual test-time adaptation (1)

Papers

Video-MMMU: Evaluating Knowledge Acquisition from Multidisciplinary Professional Videos ACL 2026 Exploring Model Kinship for Merging Large Language Models EMNLP 2025 Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss ICML 2025 BECoTTA: Input-dependent Online Blending of Experts for Continual Test-time Adaptation ICML 2024 Class-Imbalanced Graph Learning without Class Rebalancing ICML 2024 Is Self-Repair a Silver Bullet for Code Generation? ICLR 2024