{"total":69,"rows":[{"id":"2502.15603","title":"Do Multilingual LLMs Think In English?","created":"2025-02-21","cited_by_count":78,"area":"lr"},{"id":"2411.08745","title":"Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers","created":"2024-11-13","cited_by_count":44,"area":"lr"},{"id":"2506.09047","title":"Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs","created":"2025-06-10","cited_by_count":35,"area":"mllm"},{"id":"2504.11695","title":"Interpreting the linear structure of vision-language model embedding spaces","created":"2025-04-16","cited_by_count":32,"area":"mllm"},{"id":"2502.14830","title":"Middle-Layer Representation Alignment for Cross-Lingual Transfer in Fine-Tuned LLMs","created":"2025-02-20","cited_by_count":29,"area":null},{"id":"2510.04694","title":"Multilingual Routing in Mixture-of-Experts","created":"2025-10-06","cited_by_count":29,"area":"moe"},{"id":"2601.14004","title":"Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models","created":"2026-01-20","cited_by_count":27,"area":"mechinterp"},{"id":"2503.11667","title":"LogitLens4LLMs: Extending Logit Lens Analysis to Modern Large Language Models","created":"2025-02-24","cited_by_count":23,"area":"mechinterp"},{"id":"2505.13141","title":"Language-Specific Latent Process Hinders Cross-Lingual Performance","created":"2025-05-19","cited_by_count":21,"area":null},{"id":"2505.20546","title":"Paths Not Taken: Understanding and Mending the Multilingual Factual Recall Pipeline","created":"2025-05-26","cited_by_count":21,"area":null},{"id":"2506.09890","title":"The Emergence of Abstract Thought in Large Language Models Beyond Any Language","created":"2025-06-11","cited_by_count":21,"area":"lr"},{"id":"2504.10191","title":"Localized Cultural Knowledge is Conserved and Controllable in Large Language Models","created":"2025-04-14","cited_by_count":20,"area":"wm"},{"id":"2505.17407","title":"Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?","created":"2025-05-23","cited_by_count":20,"area":"lr"},{"id":"2505.15257","title":"When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners","created":"2025-05-21","cited_by_count":16,"area":"lr"},{"id":"2602.00462","title":"LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs","created":"2026-01-31","cited_by_count":15,"area":"mllm"},{"id":"2505.16415","title":"Attributing Response to Context: A Jensen-Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generation","created":"2025-05-22","cited_by_count":12,"area":"rag"},{"id":"2601.14050","title":"Understanding Multilingualism in Mixture-of-Experts LLMs: Routing Mechanism, Expert Specialization, and Layerwise Steering","created":"2026-01-20","cited_by_count":10,"area":"moe"},{"id":"2411.17666","title":"How do Multimodal Foundation Models Encode Text and Speech? An Analysis of Cross-Lingual and Cross-Modal Representations","created":"2024-11-26","cited_by_count":9,"area":null},{"id":"2505.14071","title":"Textual Steering Vectors Can Improve Visual Understanding in Multimodal Large Language Models","created":"2025-05-20","cited_by_count":9,"area":"mllm"},{"id":"2602.16660","title":"Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety Alignment","created":"2026-02-18","cited_by_count":9,"area":"alignment"},{"id":"2604.08541","title":"Seeing but Not Thinking: Routing Distraction in Multimodal Mixture-of-Experts","created":"2026-04-09","cited_by_count":9,"area":"moe"},{"id":"2505.23856","title":"OMNIGUARD: An Efficient Approach for AI Safety Moderation Across Languages and Modalities","created":"2025-05-29","cited_by_count":7,"area":"safety"},{"id":"2512.03276","title":"Too Late to Recall: Explaining the Two-Hop Problem in Multimodal Knowledge Retrieval","created":"2025-12-02","cited_by_count":7,"area":"mllm"},{"id":"2603.00510","title":"What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models","created":"2026-02-28","cited_by_count":7,"area":"mllm"},{"id":"2507.13328","title":"Vision-and-Language Training Helps Deploy Taxonomic Knowledge but Does Not Fundamentally Alter It","created":"2025-07-17","cited_by_count":6,"area":null},{"id":"2510.13796","title":"The Mechanistic Emergence of Symbol Grounding in Language Models","created":"2025-10-15","cited_by_count":6,"area":null},{"id":"2601.11778","title":"Translation as a Scalable Proxy for Multilingual Evaluation","created":"2026-01-16","cited_by_count":6,"area":null},{"id":"2603.17102","title":"Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency","created":"2026-03-17","cited_by_count":6,"area":"moe"},{"id":"2506.01074","title":"How Programming Concepts and Neurons Are Shared in Code Language Models","created":"2025-06-01","cited_by_count":5,"area":null},{"id":"2508.11536","title":"Language models align with brain regions that represent concepts across modalities","created":"2025-08-15","cited_by_count":4,"area":null},{"id":"2512.01222","title":"Unsupervised decoding of encoded reasoning using language model interpretability","created":"2025-12-01","cited_by_count":4,"area":"lr"},{"id":"2502.11364","title":"Blessing of Multilinguality: A Systematic Analysis of Multilingual In-Context Learning","created":"2025-02-17","cited_by_count":3,"area":null},{"id":"2602.01572","title":"LLM-based Embeddings: Attention Values Encode Sentence Semantics Better Than Hidden States","created":"2026-02-02","cited_by_count":3,"area":null},{"id":"2605.28042","title":"Extracting Small Translation Specialists from LLMs by Aggressively Pruning Experts","created":"2026-05-27","cited_by_count":3,"area":"moe"},{"id":"2412.12686","title":"Exploring Cross-lingual Latent Transplantation: Mutual Opportunities and Open Challenges","created":"2024-12-17","cited_by_count":2,"area":null},{"id":"2507.05387","title":"The Generalization Ridge: Information Flow in Natural Language Generation","created":"2025-07-07","cited_by_count":2,"area":null},{"id":"2509.23488","title":"Mapping Overlaps in Benchmarks through Perplexity in the Wild","created":"2025-09-27","cited_by_count":2,"area":null},{"id":"2604.20835","title":"Parallel-SFT: Improving Zero-Shot Cross-Programming-Language Transfer for Code RL","created":"2026-04-22","cited_by_count":2,"area":null},{"id":"2606.01196","title":"Low-Resource Safety Failures Are Action Failures, Not Representation Failures","created":"2026-05-31","cited_by_count":2,"area":"alignment"},{"id":"2606.26749","title":"Structure Before Collapse: Transient semantic geometry in next-token prediction","created":"2026-06-25","cited_by_count":2,"area":null},{"id":"2502.11932","title":"On Representational Dissociation of Language and Arithmetic in Large Language Models","created":"2025-02-17","cited_by_count":1,"area":null},{"id":"2509.04032","title":"What if I ask in \\textit{alia lingua}? Measuring Functional Similarity Across Languages","created":"2025-09-04","cited_by_count":1,"area":null},{"id":"2510.08284","title":"Neuron-Level Analysis of Cultural Understanding in Large Language Models","created":"2025-10-09","cited_by_count":1,"area":null},{"id":"2602.01008","title":"Adapting Where It Matters: Depth-Aware Adaptation for Efficient Multilingual Speech Recognition in Low-Resource Languages","created":"2026-02-01","cited_by_count":1,"area":null},{"id":"2603.07474","title":"Cross-Modal Taxonomic Generalization in (Vision-) Language Models","created":"2026-03-08","cited_by_count":1,"area":null},{"id":"2607.19243","title":"Inference-Time Steering for Cross-Lingual Factual Consistency in LLMs","created":"2026-07-21","cited_by_count":1,"area":"alignment"},{"id":"2609.00293","title":"Slow to See, Slow to Suppress: Understanding the Effects of Modality in Context-Memory Conflicts","created":"2026-08-31","cited_by_count":1,"area":"mllm"},{"id":"2511.04053","title":"Interpreting Multi-Attribute Confounding through Numerical Attributes in Large Language Models","created":"2025-11-06","cited_by_count":0,"area":"lr"},{"id":"2602.10504","title":"On the Robustness of Knowledge Editing for Detoxification","created":"2026-02-11","cited_by_count":0,"area":null},{"id":"2603.18863","title":"Why Better Cross-Lingual Alignment Fails for Better Cross-Lingual Transfer: Case of Encoders","created":"2026-03-19","cited_by_count":0,"area":null},{"id":"2605.09352","title":"The Wittgensteinian Representation Hypothesis: Is Language the Attractor of Multimodal Convergence?","created":"2026-05-10","cited_by_count":0,"area":null},{"id":"2606.03871","title":"Visual Instruction Tuning Aligns Modalities through Abstraction","created":"2026-06-02","cited_by_count":0,"area":null},{"id":"2606.22473","title":"Interleaved Speech Language Models Latently Work In Text","created":"2026-06-21","cited_by_count":0,"area":null},{"id":"2607.20803","title":"The Geometry of Personality: Activation Steering with Jungian Cognitive Functions","created":"2026-07-23","cited_by_count":0,"area":"mechinterp"},{"id":"2607.28969","title":"SafeNexus: Discovering and Steering Modality-Universal Safety Neurons in MLLMs","created":"2026-07-31","cited_by_count":0,"area":"mllm"},{"id":"2608.00035","title":"Linguistic Context Recodes Visual Representations in Vision-Language Models","created":"2026-07-21","cited_by_count":0,"area":"mllm"},{"id":"2608.03505","title":"ConlangBench: Exploring Language Knowledge and Learning in LLMs through Diverse Constructed Languages","created":"2026-08-04","cited_by_count":0,"area":null},{"id":"2608.15085","title":"Why Vision Fails as a Universal Bridge: Rectifying Modality Asynchrony in Multilingual MLLMs","created":"2026-08-15","cited_by_count":0,"area":"lr"},{"id":"2608.24958","title":"Can We Read the Mind of an Audio LLM? A Verbalizable, Multilingual Middle-Layer Workspace","created":"2026-08-25","cited_by_count":0,"area":"lr"},{"id":"2609.00515","title":"The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.04463","title":"Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning","created":"2026-09-03","cited_by_count":0,"area":"lr"},{"id":"2609.06381","title":"Cross-Lingual Representation Alignment by Token-Level Optimal Transport in a Language-Agnostic Space","created":"2026-09-06","cited_by_count":0,"area":null},{"id":"2609.07474","title":"Where Should Language Sit in a Multimodal Model? Lessons from What Language Does to Human Perception and Cognition","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.16967","title":"Target-Language Generation in Multilingual Models: Activation Steering and Optimal Control","created":"2026-09-15","cited_by_count":0,"area":"mechinterp"},{"id":"2609.35378","title":"Multilinguality in Hybrid Attention LLMs","created":"2026-09-28","cited_by_count":0,"area":"agents"},{"id":"2609.36205","title":"Geometric Representations of African Languages: A Regional Semantic Hub and Cultural Steering","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2609.38027","title":"Layer-Informed Fine-Tuning via Three-Stage Functional Segmentation of LLMs","created":"2026-09-28","cited_by_count":0,"area":"lr"},{"id":"2610.01921","title":"Cross-Lingual Alignment for Decoder-Only Models using MoE Routers","created":"2026-10-01","cited_by_count":0,"area":"moe"},{"id":"2610.02123","title":"Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation","created":"2026-10-01","cited_by_count":0,"area":"moe"}],"next_cursor":null}