NeuPAT: Neuron-aware Plasticity Allocation Tuning for Language-Preserving MLLMs
Jiayue Jin, Jingwei Zhang, Chen Wang, Jing Liu, Longteng Guo
cs.CL, cs.AI
Submitted: 2026-08-08
Updated: 2026-08-11
Code: https://github.com/sahil280114/codealpaca
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- GPT-4 Technical Report
- Qwen Technical Report
- Flamingo: a Visual Language Model for Few-Shot Learning
- Program Synthesis with Large Language Models
- Qwen3-VL Technical Report
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
- Training Verifiers to Solve Math Word Problems
- LLaVA-OneVision: Easy Visual Task Transfer
- Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs
- The Llama 3 Herd of Models
- Measuring Mathematical Problem Solving With the MATH Dataset
- DeepSeek-VL: Towards Real-World Vision-Language Understanding
- LLaMA: Open and Efficient Foundation Language Models
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- The Achilles' Heel of LLMs: How Altering a Handful of Neurons Can Cripple Language Abilities
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- Training-Free Mitigation of Language Reasoning Degradation After Multimodal Instruction Tuning
- PlaM: Training-Free Plateau-Guided Model Merging for Better Visual Grounding in MLLMs
- 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