Model Merging

FusionBench: A Unified Library and Comprehensive Benchmark for Deep Model Fusion JMLR 2025 FW-Merging: Scaling Model Merging with Frank-Wolfe Optimization ICCV 2025 GenKnowSub: Improving Modularity and Reusability of LLMs through General Knowledge Subtraction ACL 2025 GeoEdit: Geometric Knowledge Editing for Large Language Models EMNLP 2025 Glider: Global and Local Instruction-Driven Expert Router EMNLP 2025 GLoCIM: Global-view Long Chain Interest Modeling for news recommendation COLING 2025 GT-NLP at SemEval-2025 Task 11: EmoRationale, Evidence-Based Emotion Detection via Retrieval-Augmented Generation SemEval 2025 Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generation EMNLP 2025 Heterogeneous Federated Learning with Scalable Server Mixture-of-Experts IJCAI 2025 How to Merge Your Multimodal Models Over Time? CVPR 2025 HydraOpt: Navigating the Efficiency-Performance Trade-off of Adapter Merging EMNLP 2025 ImPart: Importance-Aware Delta-Sparsification for Improved Model Compression and Merging in LLMs ACL 2025 Improve LLM-as-a-Judge Ability as a General Ability EMNLP 2025 In-Context Meta LoRA Generation IJCAI 2025 Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planning ICCV 2025 IterIS: Iterative Inference-Solving Alignment for LoRA Merging CVPR 2025 LaMDAgent: An Autonomous Framework for Post-Training Pipeline Optimization via LLM Agents EMNLP 2025 Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge EMNLP 2025 Layer-Level Self-Exposure and Patch: Affirmative Token Mitigation for Jailbreak Attack Defense NAACL 2025 Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models ICLR 2025 Learning Compatible Multi-Prize Subnetworks for Asymmetric Retrieval CVPR 2025 Learning from Hallucinations: Mitigating Hallucinations in LLMs via Internal Representation Intervention IJCNLP 2025 Learning Harmonized Representations for Speculative Sampling ICLR 2025 LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint ACL 2025 Less for More: Enhanced Feedback-aligned Mixed LLMs for Molecule Caption Generation and Fine-Grained NLI Evaluation ACL 2025