# MANTA: Cross-Modal Semantic Alignment and Information-Theoretic Optimization for Long-form Multimodal Understanding

arXiv: https://arxiv.org/abs/2507.00068 · HTML: https://mlsanity.com/paper/2507.00068/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 36 references that are in the map.
- First posted on arXiv 28 Jun 2025.

## Builds on (most cited first, 16 of 16)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality](https://mlsanity.com/paper/2304.14178/) (2023)
- [InternVideo: General Video Foundation Models via Generative and Discriminative Learning](https://mlsanity.com/paper/2212.03191/) (2022)
- [TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding](https://mlsanity.com/paper/2312.02051/) (2023)
- [Long-Term Feature Banks for Detailed Video Understanding](https://mlsanity.com/paper/1812.05038/) (2018)
- [Zero-Shot Video Question Answering via Frozen Bidirectional Language Models](https://mlsanity.com/paper/2206.08155/) (2022)
- [Video-XL: Extra-Long Vision Language Model for Hour-Scale Video Understanding](https://mlsanity.com/paper/2409.14485/) (2024)
- [VIOLET : End-to-End Video-Language Transformers with Masked Visual-token Modeling](https://mlsanity.com/paper/2111.12681/) (2021)
- [A Simple LLM Framework for Long-Range Video Question-Answering](https://mlsanity.com/paper/2312.17235/) (2023)
- [D4: Improving LLM Pretraining via Document De-Duplication and Diversification](https://mlsanity.com/paper/2308.12284/) (2023)
- [ST-LLM: Large Language Models Are Effective Temporal Learners](https://mlsanity.com/paper/2404.00308/) (2024)
- [Momentor: Advancing Video Large Language Model with Fine-Grained Temporal Reasoning](https://mlsanity.com/paper/2402.11435/) (2024)
- [Retrieval Augmented Visual Question Answering with Outside Knowledge](https://mlsanity.com/paper/2210.03809/) (2022)
- [PreFLMR: Scaling Up Fine-Grained Late-Interaction Multi-modal Retrievers](https://mlsanity.com/paper/2402.08327/) (2024)
- [Filling the Image Information Gap for VQA: Prompting Large Language Models to Proactively Ask Questions](https://mlsanity.com/paper/2311.11598/) (2023)

## Abstract (arXiv)

While multi-modal learning has advanced significantly, current approaches often treat modalities separately, creating inconsistencies in representation and reasoning. We introduce MANTA (Multi-modal Abstraction and Normalization via Textual Alignment), a theoretically-grounded framework that unifies visual and auditory inputs into a structured textual space for seamless processing with large language models. MANTA addresses four key challenges: (1) semantic alignment across modalities with information-theoretic optimization, (2) adaptive temporal synchronization for varying information densities, (3) hierarchical content representation for multi-scale understanding, and (4) context-aware retrieval of sparse information from long sequences. We formalize our approach within a rigorous mathematical framework, proving its optimality for context selection under token constraints. Extensive experiments on the challenging task of Long Video Question Answering show that MANTA improves state-of-the-art models by up to 22.6% in overall accuracy, with particularly significant gains (27.3%) on videos exceeding 30 minutes. Additionally, we demonstrate MANTA's superiority on temporal reasoning tasks (23.8% improvement) and cross-modal understanding (25.1% improvement). Our framework introduces novel density estimation techniques for redundancy minimization while preserving rare signals, establishing new foundations for unifying multimodal representations through structured text.