arXiv AI

AC-JEPA: Anchored Cross-Modal Joint Embedding Predictive Architecture for Visual-to-Symbolic Translation

AC-JEPA is a plug‑in for supervised fine‑tuning that predicts masked intermediate states in both directions of a cross‑modal translation task, combining a joint‑embedding objective with causal token‑level supervision. It is designed to be robust to imperfect paired data by learning from continuous latent targets derived from pseudo‑labels, rather than hard categorical supervision. In optical chemical structure recognition, AC‑JEPA outperforms plain supervised fine‑tuning, achieving 76.07 % accuracy on clean pairs and 80.23 % when self‑training with pseudo‑labels, whereas SFT alone drops to 63.47 % under the same conditions.

arXiv AI

Oct 16

MSP: Modality Self-Play for Training Multimodal LLMs Without Paired Cross-Modal Data

The paper introduces Modality Self-Play (MSP), a training framework that enables multimodal large language models to learn cross‑modal interactions without requiring paired non‑text data. MSP uses proxy‑mediated textual descriptions of unseen modalities as semantic bridges, allowing a model to generate and align these proxies with observed modalities. Experiments show that MSP achieves state‑of‑the‑art or competitive performance on several multimodal benchmarks and demonstrates effective zero‑shot modality composition through learned bridges and anchor‑supporting proxy descriptions.

By Robert Moseley, Md Kaykobad Reza, Ameya Patil, Edward Ayrapetian, Salman Asif

llms reinforcement-learning multimodal benchmarks safety

arXiv Computation and Language

Oct 12

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs

TokenSwap introduces a method to benchmark and reduce the modality gap in multimodal large language models (MLLMs) by interleaving semantically aligned images with text tokens. By transforming text-only benchmarks like MMLU into image‑interleaved versions (TokenSwap‑Bench), the authors reveal a pervasive performance drop—averaging 19.6%—across 42 MLLMs, with reasoning models showing smaller gaps. Training with TokenSwap mitigates this gap while maintaining strong text‑only and vision‑language performance.

By Andong Hua, Colton Bishop, Igor Mordatch, Arian Hosseini, Jindong Gu, Aleksandra Faust, Rebecca Roelofs, Yao Qin

llms multimodal benchmarks safety

arXiv Machine Learning

Jul 13

Mixture of Probes: Learning from Privileged Modalities in Multimodal LLMs Through Probing

arXiv:2607. 08839v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) are typically designed under the assumption that all modalities available during training will also be accessible at inference.

By Dominick Reilly, Qiyu Wu, Hiromi Wakaki, Srijan Das, Yuki Mistufuji

llms multimodal safety

arXiv Machine Learning

Jun 10

When to Align, When to Predict: A Phase Diagram for Multimodal Learning

arXiv:2606. 11190v1 Announce Type: new Abstract: Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each succeeds, when each fails, and when cross-modal training helps at all -- a gap that leaves practitioners, especially in scientific domains like biomedicine or astrophysics, with heterogeneous instruments and multiple levels of organization and measurement, unable to diagnose why standard methods underperform the best single modality.

By Ilay Kamai, Hugues Van Assel, Aviv Regev, Hagai B. Perets, Randall Balestriero

multimodal benchmarks safety

arXiv AI

Jun 17

Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model

arXiv:2606. 17950v1 Announce Type: cross Abstract: Visual information helps resolve ambiguity in coreference resolution, leading to notable performance gains.

By Jinghan Wu, Jing Li, Ivor W. Tsang, Xuetao Zhang

multimodal benchmarks safety

arXiv Machine Learning

Sep 16

Towards Scalable RLVR: Multimodal Instruction Following Data Synthesis and Distillation

arXiv:2609.16059v1 Announce Type: cross Abstract: Multimodal instruction following (MMIF) is crucial for building generalist agents. However, current training paradigms rely heavily on Supervised Fin...

By Yirong Zeng, Zhang Sai, Yuxian Wang, Yutai Hou, Yufei Liu, Xiao Ding, Bibo Cai

agents reinforcement-learning fine-tuning efficiency multimodal benchmarks

Hugging Face Trending Papers

Jun 9

When to Align, When to Predict: A Phase Diagram for Multimodal Learning

Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each succeeds, when each fails, and when cross-modal training helps at all -- a gap that leaves practitioners, especially in scientific domains like biomedicine or astrophysics, with heterogeneous instruments and multiple levels of organization and measurement, unable to diagnose why standard methods underperform the best single modality. We develop a unified linear framework that addresses both questions.

multimodal benchmarks safety

arXiv AI

Jun 24

Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations

arXiv:2606. 24716v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) are increasingly used to extract interpretable concepts from vision and vision language models, yet existing evaluation methods largely rely on proxy metrics or qualitative inspection rather than measuring semantic correspondence.

By Jonas Klotz, Cassio F. Dantas, Pallavi Jain, Diego Marcos, Beg\"um Demir

llms rag benchmarks safety

arXiv AI

Oct 7

Dynamic Alignment and Calibration for Multimodal Learning

arXiv:2610.07928v1 Announce Type: cross Abstract: Dynamic multimodal learning aims to learn robust representations by adaptively modeling information discrepancies across modalities. However, existin...

By Jinghao Xu, Zhenhua Guo, Xiaofeng Zhu, Xiaoshuang Shi

multimodal benchmarks safety

arXiv Computer Vision

Sep 23

0.5%>100%: Bidirectional Reciprocal Learning for Referring Image Segmentation

arXiv:2609.24510v2 Announce Type: replace Abstract: Recent advances in vision foundation models (VFMs) have shown remarkable capabilities across diverse unimodal visual tasks. However, adapting VFMs...

By Xiaoqiang Lu, Licheng Jiao, Lingling Li, Yuting Yang, Long Sun, Wenping Ma, Xu Liu, Fang Liu

computer-vision fine-tuning multimodal benchmarks safety

arXiv AI

Jun 8

Modality Gap-Driven Subspace Alignment Training Paradigm For Multimodal Large Language Models

arXiv:2602. 07026v3 Announce Type: replace-cross Abstract: Despite the success of multimodal contrastive learning in aligning visual and linguistic representations, a persistent geometric anomaly, the Modality Gap, remains: embeddings of distinct modalities expressing identical semantics occupy systematically offset regions.

By Xiaomin Yu, Yi Xin, Yuhui Zhang, Wenjie Zhang, Chonghan Liu, Hanzhen Zhao, Chen Liu, Xiaoxing Hu, Ziyue Qiao, Hao Tang, Xiaobin Hu, Chengwei Qin, Hui Xiong, Yu Qiao, Shuicheng Yan

llms rag fine-tuning multimodal safety

arXiv Machine Learning

Jun 25

MJEPA: A Simple and Scalable Joint-Embedding Predictive Architecture for Audio-Visual Learning

arXiv:2606. 25225v1 Announce Type: cross Abstract: Self-supervised learning from large-scale video data has emerged as a dominant paradigm for visual representation learning.

By Revant Teotia, Adrien Bardes, Michael Rabbat, Sumit Chopra, Matthew J. Muckley, Nicolas Ballas

rag multimodal benchmarks

Related stories