Suyuchen Wang
  • Bio
  • Papers
  • Experience
  • Projects
  • Teaching
  • Projects
  • Experience
  • Blog
  • Publications
    • System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts
    • Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems
    • AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Understanding
    • BigDocs: An Open Dataset for Training Multimodal Models on Document and Code Tasks
    • CARE: Improving Context Fidelity via Native Retrieval-Augmented Reasoning
    • FinSage: A Multi-aspect RAG System for Financial Filings Question Answering
    • GraphOmni: A Comprehensive and Extendable Benchmark Framework for Large Language Models on Graph-theoretic Tasks
    • LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models
    • MAP: Low-compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation
    • R(^mbox3)Mem: Bridging Memory Retention and Retrieval via Reversible Compression
    • R³Mem: Bridging Memory Retention and Retrieval via Reversible Compression
    • Rethinking Decentralized Learning: Towards More Realistic Evaluations with a Metadata-Agnostic Approach
    • STRICT: Stress-Test of Rendering Image Containing Text
    • VCR: Pixel-Level Complex Reasoning by Restoring Occluded Text
    • WebMMU: A Benchmark for Multimodal Multilingual Website Understanding and Code Generation
    • FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact Retrieval
    • LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models
    • Resonance RoPE: Improving Context Length Generalization of Large Language Models
    • DiffTune: A Diffusion-Based Approach to Diverse Instruction-Tuning Data Generation
    • Efficient Classification of Long Documents via State-Space Models
    • Multimodal Multi-Hop Question Answering Through a Conversation Between Tools and Efficiently Finetuned Large Language Models
    • QEN: Applicable Taxonomy Completion via Evaluating Full Taxonomic Relations
    • Enquire One's Parent and Child Before Decision: Fully Exploit Hierarchical Structure for Self-Supervised Taxonomy Expansion
    • Anchored Model Transfer and Soft Instance Transfer for Cross-Task Cross-Domain Learning: A Study Through Aspect-Level Sentiment Classification
  • Recent & Upcoming Talks
    • Example Talk
  • Teaching
    • IFT 6289 - Winter 2024
    • IFT 6759 - Winter 2023
    • IFT 6289 - Winter 2022

MAP: Low-compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation

Jan 1, 2025·
Lu Li
,
Tianyu Zhang
,
Zhiqi Bu
,
Suyuchen Wang
,
Huan He
,
Jie Fu
,
Yonghui Wu
,
Jiang Bian
,
Yong Chen
,
Yoshua Bengio
· 0 min read
Cite URL
Type
Conference paper
Last updated on Apr 20, 2025
← LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models Jan 1, 2025
R(^mbox3)Mem: Bridging Memory Retention and Retrieval via Reversible Compression Jan 1, 2025 →

© 2026 Suyuchen Wang. This webpage is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.