Image de couverture de livre

Compose and Fuse: Revisiting the Foundational Bottlenecks in Multimodal Reasoning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Wang, Yucheng, Hou, Yifan, Javadov, Aydin, Akhtar, Mubashara, Sachan, Mrinmaya
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Chimera: Diagnosing Shortcut Learning in Visual-Language Understanding
par: Chi, Ziheng, et autres
Publié: (2025)

Benchmarking and Enhancing Text-to-Image Models for Generating Visual Representations in Early Arithmetic Education
par: Wang, Junling, et autres
Publié: (2026)

Probing for Arithmetic Errors in Language Models
par: Sun, Yucheng, et autres
Publié: (2025)

ReProbe: Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States of Large Language Models
par: Ni, Jingwei, et autres
Publié: (2025)

Can Vision-Language Models Solve Visual Math Equations?
par: Choudhury, Monjoy Narayan, et autres
Publié: (2025)

Corrupted by Reasoning: Reasoning Language Models Become Free-Riders in Public Goods Games
par: Piedrahita, David Guzman, et autres
Publié: (2025)

How to Engage Your Readers? Generating Guiding Questions to Promote Active Reading
par: Cui, Peng, et autres
Publié: (2024)

Simulating Students or Sycophantic Problem Solving? On Misconception Faithfulness of LLM Simulators
par: Do, Heejin, et autres
Publié: (2026)

Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors
par: Daheim, Nico, et autres
Publié: (2024)

Ev2R: Evaluating Evidence Retrieval in Automated Fact-Checking
par: Akhtar, Mubashara, et autres
Publié: (2024)

Generating Pedagogically Meaningful Visuals for Math Word Problems: A New Benchmark and Analysis of Text-to-Image Models
par: Wang, Junling, et autres
Publié: (2025)

Learning to Reason Efficiently with A* Post-Training
par: Opedal, Andreas, et autres
Publié: (2026)

Improving Large Language Model Safety with Contrastive Representation Learning
par: Simko, Samuel, et autres
Publié: (2025)

Can LLMs Model Incorrect Student Reasoning? A Case Study on Distractor Generation
par: Zengaffinen, Yanick, et autres
Publié: (2026)

Multilingual Performance Biases of Large Language Models in Education
par: Gupta, Vansh, et autres
Publié: (2025)

Towards Aligning Language Models with Textual Feedback
par: Lloret, Saüc Abadal, et autres
Publié: (2024)

Can Reasoning Help Large Language Models Capture Human Annotator Disagreement?
par: Ni, Jingwei, et autres
Publié: (2025)

On the Emergence of Induction Heads for In-Context Learning
par: Musat, Tiberiu, et autres
Publié: (2025)

AFaCTA: Assisting the Annotation of Factual Claim Detection with Reliable LLM Annotators
par: Ni, Jingwei, et autres
Publié: (2024)

Post-Training Language Models for Crosslingual Consistency
par: Liu, Tianyu, et autres
Publié: (2026)

Retro-li: Small-Scale Retrieval Augmented Generation Supporting Noisy Similarity Searches and Domain Shift Generalization
par: Rashiti, Gentiana, et autres
Publié: (2024)

MathGAP: Out-of-Distribution Evaluation on Problems with Arbitrarily Complex Proofs
par: Opedal, Andreas, et autres
Publié: (2024)

From Problem-Solving to Teaching Problem-Solving: Aligning LLMs with Pedagogy using Reinforcement Learning
par: Dinucu-Jianu, David, et autres
Publié: (2025)

LEXam: Benchmarking Legal Reasoning on 340 Law Exams
par: Fan, Yu, et autres
Publié: (2025)

PATS: Personality-Aware Teaching Strategies with Large Language Model Tutors
par: Rooein, Donya, et autres
Publié: (2026)

MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors
par: Macina, Jakub, et autres
Publié: (2025)

Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation
par: Liu, Tianyu, et autres
Publié: (2024)

DIRAS: Efficient LLM Annotation of Document Relevance in Retrieval Augmented Generation
par: Ni, Jingwei, et autres
Publié: (2024)

Are Language Models Efficient Reasoners? A Perspective from Logic Programming
par: Opedal, Andreas, et autres
Publié: (2025)

Exploring the Jungle of Bias: Political Bias Attribution in Language Models via Dependency Analysis
par: Jenny, David F., et autres
Publié: (2023)

Balancing Truthfulness and Informativeness with Uncertainty-Aware Instruction Fine-Tuning
par: Wu, Tianyi, et autres
Publié: (2025)

Dense SAE Latents Are Features, Not Bugs
par: Sun, Xiaoqing, et autres
Publié: (2025)

Confidence Regulation Neurons in Language Models
par: Stolfo, Alessandro, et autres
Publié: (2024)

Autoformalizing Natural Language to First-Order Logic: A Case Study in Logical Fallacy Detection
par: Lalwani, Abhinav, et autres
Publié: (2024)

CLadder: Assessing Causal Reasoning in Language Models
par: Jin, Zhijing, et autres
Publié: (2023)

Do Vision-Language Models Really Understand Visual Language?
par: Hou, Yifan, et autres
Publié: (2024)

What Do Language Models Learn in Context? The Structured Task Hypothesis
par: Li, Jiaoda, et autres
Publié: (2024)

Date Fragments: A Hidden Bottleneck of Tokenization for Temporal Reasoning
par: Bhatia, Gagan, et autres
Publié: (2025)

Revisiting Model Interpolation for Efficient Reasoning
par: Wu, Taiqiang, et autres
Publié: (2025)

Augmenting Math Word Problems via Iterative Question Composing
par: Liu, Haoxiong, et autres
Publié: (2024)