Cover Image

Mechanistic Interpretability of Large-Scale Counting in LLMs through a System-2 Strategy

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hasani, Hosein, Banayeeanzade, Mohammadali, Nafisi, Ali, Mohammadian, Sadegh, Askari, Fatemeh, Bagherian, Mobin, Izadi, Amirmohammad, Baghshah, Mahdieh Soleymani
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Understanding Counting Mechanisms in Large Language and Vision-Language Models
by: Hasani, Hosein, et al.
Published: (2025)

Uncovering Grounding IDs: How External Cues Shape Multimodal Binding
by: Hasani, Hosein, et al.
Published: (2025)

Causal Attribution via Activation Patching
by: Izadi, Amirmohammad, et al.
Published: (2026)

Visual Structures Helps Visual Reasoning: Addressing the Binding Problem in VLMs
by: Izadi, Amirmohammad, et al.
Published: (2025)

CAREL: Instruction-guided reinforcement learning with cross-modal auxiliary objectives
by: Saghafian, Armin, et al.
Published: (2024)

Attention Overlap Is Responsible for The Entity Missing Problem in Text-to-image Diffusion Models!
by: Marioriyad, Arash, et al.
Published: (2024)

Limits and Gains of Test-Time Scaling in Vision-Language Reasoning
by: Ahmadpour, Mohammadjavad, et al.
Published: (2025)

Analyzing CLIP's Performance Limitations in Multi-Object Scenarios: A Controlled High-Resolution Study
by: Abbasi, Reza, et al.
Published: (2025)

CLIP Under the Microscope: A Fine-Grained Analysis of Multi-Object Representation
by: Abbasi, Reza, et al.
Published: (2025)

Improving 3D Few-Shot Segmentation with Inference-Time Pseudo-Labeling
by: Mozafari, Mohammad, et al.
Published: (2024)

HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models
by: Zohrabi, Reihaneh, et al.
Published: (2026)

Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
by: Zohrabi, Reihaneh, et al.
Published: (2025)

ComAlign: Compositional Alignment in Vision-Language Models
by: Abdollah, Ali, et al.
Published: (2024)

CER: Confidence Enhanced Reasoning in LLMs
by: Razghandi, Ali, et al.
Published: (2025)

The Illusion of Procedural Reasoning: Measuring Long-Horizon FSM Execution in LLMs
by: Samiei, Mahdi, et al.
Published: (2025)

SUSD: Structured Unsupervised Skill Discovery through State Factorization
by: Hosseini, Seyed Mohammad Hadi, et al.
Published: (2026)

Dilated Balanced Cross Entropy Loss for Medical Image Segmentation
by: Hosseini, Seyed Mohsen, et al.
Published: (2024)

Classification of Breast Cancer Histopathology Images using a Modified Supervised Contrastive Learning Method
by: Sani, Matina Mahdizadeh, et al.
Published: (2024)

PreND: Enhancing Intrinsic Motivation in Reinforcement Learning through Pre-trained Network Distillation
by: Davoodabadi, Mohammadamin, et al.
Published: (2024)

Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing
by: Marioriyad, Arash, et al.
Published: (2026)

T2I-FineEval: Fine-Grained Compositional Metric for Text-to-Image Evaluation
by: Hosseini, Seyed Mohammad Hadi, et al.
Published: (2025)

LLM-Agent-Controller: A Universal Multi-Agent Large Language Model System as a Control Engineer
by: Zahedifar, Rasoul, et al.
Published: (2025)

Eye-Q: A Multilingual Benchmark for Visual Word Puzzle Solving and Image-to-Phrase Reasoning
by: Najar, Ali, et al.
Published: (2026)

Large Language Models for Scientific Idea Generation: A Creativity-Centered Survey
by: Shahhosseini, Fatemeh, et al.
Published: (2025)

The Silent Judge: Unacknowledged Shortcut Bias in LLM-as-a-Judge
by: Marioriyad, Arash, et al.
Published: (2025)

LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions
by: Mehri, Faridoun, et al.
Published: (2024)

Deciphering the Role of Representation Disentanglement: Investigating Compositional Generalization in CLIP Models
by: Abbasi, Reza, et al.
Published: (2024)

SOInter: A Novel Deep Energy Based Interpretation Method for Explaining Structured Output Models
by: Seyyedsalehi, S. Fatemeh, et al.
Published: (2022)

Trained Models Tell Us How to Make Them Robust to Spurious Correlation without Group Annotation
by: Ghaznavi, Mahdi, et al.
Published: (2024)

Fine-Grained Alignment and Noise Refinement for Compositional Text-to-Image Generation
by: Izadi, Amir Mohammad, et al.
Published: (2025)

Efficient Adversarial Attacks on High-dimensional Offline Bandits
by: Hosseini, Seyed Mohammad Hadi, et al.
Published: (2026)

VQEL: Enabling Self-Play in Emergent Language Games via Agent-Internal Vector Quantization
by: Paqaleh, Mohammad Mahdi Samiei, et al.
Published: (2025)

Persian MusicGen: A Large-Scale Dataset and Culturally-Aware Generative Model for Persian Music
by: Sameti, Mohammad Hossein, et al.
Published: (2026)

Infinity and Beyond: Compositional Alignment in VAR and Diffusion T2I Models
by: Shahabadi, Hossein, et al.
Published: (2025)

Language Plays a Pivotal Role in the Object-Attribute Compositional Generalization of CLIP
by: Abbasi, Reza, et al.
Published: (2024)

Why Settle for Mid: A Probabilistic Viewpoint to Spatial Relationship Alignment in Text-to-image Models
by: Rezaei, Parham, et al.
Published: (2025)

Diffusion Beats Autoregressive: An Evaluation of Compositional Generation in Text-to-Image Models
by: Marioriyad, Arash, et al.
Published: (2024)

Mechanistic Interpretability of Emotion Inference in Large Language Models
by: Tak, Ala N., et al.
Published: (2025)

Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generation
by: Abootorabi, Mohammad Mahdi, et al.
Published: (2025)

The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation
by: Marioriyad, Arash, et al.
Published: (2026)