# Scaling-Aware Adapter for Structure-Grounded LLM Reasoning

arXiv: https://arxiv.org/abs/2602.02780 · HTML: https://mlsanity.com/paper/2602.02780/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 28 of its 70 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 28 of 28)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Translation between Molecules and Natural Language](https://mlsanity.com/paper/2204.11817/) (2022)
- [Mol-Instructions: A Large-Scale Biomolecular Instruction Dataset for Large Language Models](https://mlsanity.com/paper/2306.08018/) (2023)
- [A Molecular Multimodal Foundation Model Associating Molecule Graphs with Natural Language](https://mlsanity.com/paper/2209.05481/) (2022)
- [MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter](https://mlsanity.com/paper/2310.12798/) (2023)
- [Ministral 3](https://mlsanity.com/paper/2601.08584/) (2026)
- [All-atom Diffusion Transformers: Unified generative modelling of molecules and materials](https://mlsanity.com/paper/2503.03965/) (2025)
- [Towards 3D Molecule-Text Interpretation in Language Models](https://mlsanity.com/paper/2401.13923/) (2024)
- [Dynamic Parallel Tree Search for Efficient LLM Reasoning](https://mlsanity.com/paper/2502.16235/) (2025)
- [ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding](https://mlsanity.com/paper/2408.11363/) (2024)
- [ProtChatGPT: Towards Understanding Proteins with Large Language Models](https://mlsanity.com/paper/2402.09649/) (2024)
- [MatterChat: A Multi-Modal LLM for Material Science](https://mlsanity.com/paper/2502.13107/) (2025)
- [Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model](https://mlsanity.com/paper/2502.13449/) (2025)
- [Baize: An Open-Source Chat Model with Parameter-Efficient Tuning on Self-Chat Data](https://mlsanity.com/paper/2304.01196/) (2023)
- [3D-MolT5: Leveraging Discrete Structural Information for Molecule-Text Modeling](https://mlsanity.com/paper/2406.05797/) (2024)
- [RNA-GPT: Multimodal Generative System for RNA Sequence Understanding](https://mlsanity.com/paper/2411.08900/) (2024)
- [Prot2Text-V2: Protein Function Prediction with Multimodal Contrastive Alignment](https://mlsanity.com/paper/2505.11194/) (2025)
- [Chem3DLLM: 3D Multimodal Large Language Models for Chemistry](https://mlsanity.com/paper/2508.10696/) (2025)
- [MolGraph-xLSTM: A graph-based dual-level xLSTM framework with multi-head mixture-of-experts for enhanced molecular representation and interpretability](https://mlsanity.com/paper/2501.18439/) (2025)
- [Structure-Aware Fusion with Progressive Injection for Multimodal Molecular Representation Learning](https://mlsanity.com/paper/2510.23640/) (2025)
- [LLM3-DTI: A Large Language Model and Multi-modal data co-powered framework for Drug-Target Interaction prediction](https://mlsanity.com/paper/2511.06269/) (2025)

## Abstract (arXiv)

Large language models (LLMs) are enabling reasoning over 2D and 3D structures, yet existing methods remain modality-specific and typically compress structural inputs through sequence-based tokenization or fixed-length query connectors. Such architectures either omit the geometric grounding requisite for mitigating structural hallucinations, or impose inflexible modality fusion bottlenecks that concurrently over-compress and suboptimally allocate structural tokens, thereby impeding the realization of generalized all-atom reasoning. We introduce Cuttlefish, a unified multimodal LLM that grounds language reasoning in geometric cues while scaling modality tokens with structural complexity. First, Scaling-Aware Patching leverages an instruction-conditioned gating mechanism to generate variable-size patches over structural graphs, adaptively scaling the query token budget with structural complexity to mitigate fixed-length connector bottlenecks. Second, Geometry Grounding Adapter refines these adaptive tokens via cross-attention to modality embeddings and injects the resulting modality tokens into the LLM, exposing explicit geometric cues to reduce structural hallucination. Experiments across interdisciplinary all-atom benchmarks demonstrate that Cuttlefish achieves superior performance in heterogeneous structure-grounded reasoning. Code: github.com/zihao-jing/Cuttlefish.