ResMerge: Residual-based Spectral Merging of Large Language Models

arXiv:2606.02252 · cs.CL · Submitted 2026-06-01 · Read on arXiv

cs.CL

Submitted: 2026-06-01

Updated: 2026-08-26

Comments: 19 pages including appendix. Accepted to the EMNLP 2026 Main Conference

Code: https://github.com/sunyd0303-cpu/ResMerge-release

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 23× in this paper · explained in 2 episodes
• Model Merging — 17× in this paper · explained in 2 episodes
• Alignment — 16× in this paper · explained in 1 episode
• Ablation — 14× in this paper · explained in 1 episode
• Aggregation — 9× in this paper · explained in 1 episode
• Adaptation — 7× in this paper · explained in 1 episode
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Reinforcement Learning — 6× in this paper · explained in 4 episodes
• Lightweight Head Correction — 5× in this paper
• Consistency — 4× in this paper · explained in 3 episodes
• Accuracy — 3× in this paper · explained in 1 episode
• Embeddings — 3× in this paper · explained in 2 episodes
• GRPO — 3× in this paper · explained in 6 episodes
• Reinforcement Learning (RL) — 3× in this paper · explained in 25 episodes
• Singular Value Decomposition (SVD) — 3× in this paper · explained in 2 episodes
• Ablation Studies — 2× in this paper · explained in 1 episode
• Alignment Tax — 2× in this paper · explained in 4 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Calibration — 1× in this paper · explained in 7 episodes
• Pruning — 1× in this paper · explained in 4 episodes
• Redundancy — 1× in this paper · explained in 2 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model
• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
• AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
• Program Synthesis with Large Language Models
• When Shared Knowledge Hurts: Spectral Over-Accumulation in Model Merging
• SALSA: Soup-based Alignment Learning for Stronger Adaptation in RLHF
• Let's Verify Step by Step
• Evaluating Large Language Models Trained on Code
• Task Singular Vectors: Reducing Task Interference in Model Merging
• General-Reasoner: Advancing LLM Reasoning Across All Domains
• Qwen2.5 Technical Report
• No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces
• Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
• Behavior Knowledge Merge in Reinforced Agentic Models
• SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild
• WARP: On the Benefits of Weight Averaged Rewarded Policies
• GPQA: A Graduate-Level Google-Proof Q&A Benchmark

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering