Concept Alignment

Latest papers

13 papers2 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Language Unalignability: Why Some Concepts Resist Cross-Cultural Benchmark Evaluation
Oct 6, 2026Shu-Kai Hsieh, Da-Chen LianCultural Knowledge in Language Models LLM Evaluation
National Taiwan University
Alignment Games: A Framework for Conceptual Repair in Human-AI Collaboration
Sep 28, 2026Hari Subramonyam, Maneesh Agrawala, Sean FollmerConcept Alignment Human-AI Collaboration
Stanford University USA
ITL: Interpretable Document Alignment with Structured Reference Frameworks
Aug 27, 2026Raúl Giráldez, Dayrelis Mena, Jesús S. Aguilar--RuizSemantic Alignment Concept Alignment
School of Engineering Pablo de Olavide University ES-41013 Seville, Spain
Convergence in Science, Divergence in Religion: Calibrated Framing Differences Across Wikipedia's Language Editions
Aug 22, 2026Hung-Hsuan ChenConcept Alignment Cross-Lingual Representation Alignment
Computer Science and Information Engineering National Central University Taoyuan, Taiwan
Semantic Alignment of AI Models: Concept Collapse, Checkpoint Dynamics, and Cross-Lingual Transfer
Aug 3, 2026Tyler Ashoff, Jordan RoduMultilingual Language Model Evaluation Representation Geometry in Language Models
Department of Statistics, University of Virginia. · Department of Statistics, University of Virginia
Forgetting Our Way to Shared Meaning: Effects of Forgetting on Conceptual Alignment in a Non-Partnership Coordination Game
Jul 13, 2026Landon Liu, Mary Kelly, Alan TsangAgent Memory Game Theory
Department of Philosophy, Carnegie Mellon University · Department of Cognitive Science, Carleton University · Department of Computer Science, Carleton University
A Taxonomy of Conceptual Alignment in Human-Robot Dialogue
Jun 21, 2026Shengchen Zhang, Xiaohua Sun, Weiwei GuoSpoken Dialogue Systems Concept Alignment
College of Design and Innovation, Tongji University, Shanghai, China. · School of Design, Southern University of Science and Technology, Shenzhen, Guangdong, China.
A Unifying Framework for Concept-Based Representational Similarity
Jun 8, 2026Grégoire Dhimoïla, Agustin Martin Picard, Victor Boutin +2Representational Similarity Analysis Cross-Modal Alignment
Brown University · ENS Paris Saclay · DEEL - IRT Saint Exup´ery +2
Investigating Concept Alignment Using Implausible Category Members
May 20, 2026Sunayana Rane, Brenden M. Lake, Thomas L. GriffithsAI Alignment Concept Alignment
Department of Computer Science · Princeton University · Princeton, NJ 08540 +1
XSearch: Explainable Code Search via Concept-to-Code Alignment
May 15, 2026Yiming Liu, Ruofan Liu, Yun Lin +7Concept Alignment OOD Generalization
Shanghai Jiao Tong University, China · Shanghai Innovation Institute, China · National University of Singapore, Singapore +3
Prototype-Grounded Concept Models for Verifiable Concept Alignment
Apr 17, 2026Stefano Colamonaco, David Debot, Pietro Barbiero +1Concept Bottleneck Models Concept Alignment
Department of Computer Science, KU Leuven · IBM Research, Zurich
Training-Free Image Editing with Visual Context Integration and Concept Alignment
Apr 6, 2026Rui Song, Guo-Hua Wang, Qing-Guo Chen +6Concept Alignment Training-Free Image Editing
HKUST · Alibaba International Digital Commerce Group · Institute for AI Industry Research (AIR), Tsinghua University +1
Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models
Mar 31, 2026Christine Zhang, Dan Jurafsky, Chen ShaniLanguage Modeling LLM Alignment
Stanford University

Oct 6, 2026J/K move · Enter open · S save

Language Unalignability: Why Some Concepts Resist Cross-Cultural Benchmark Evaluation

Shu-Kai Hsieh, Da-Chen Lian

National Taiwan University

Cultural Knowledge in Language Models LLM Evaluation Cross-Cultural Language Model Evaluation Multilingual Language Model Evaluation Cross-Lingual Representation Alignment Concept Alignment+1

Open paper AskPDF arXiv

Current evaluation of multilingual Large Language Models (LLMs) rests on an implicit Translation-Isomorphism Assumption (TIA): that semantic structures across languages are congruent and mutually mappable without loss of information. We argue that this assumption is not merely violated in practice, but ill-posed in principle for a typologically identifiable class of concepts, including pragmatic markers, honorifics, and diachronically stratified terms. We formalize this failure using a usage-cloud framework, representing concepts as point sets of contextualized embeddings. We define

α

-unalignability as the impossibility of any mapping that simultaneously preserves lexical faithfulness (centroid correspondence) and structural faithfulness (local neighborhood topology). We provide three layers of evidence. Behaviorally, we show that FLORES-200 translation failures are predicted by language family and resource class but not by script, and that LOBSTER reasoning scores vary by family. Mechanistically, we report a Representation-Intervention Gap (RIG) in a nine-model case study on Yami: the models' activations encode a regularity along which Yami groups with other low-resource and Austronesian languages, yet interventions on language-specific neurons show no demonstrated advantage over random masks: the regularity is visible but not usable by this intervention. Finally, we operationalize these findings into a multidimensional diagnostic profile: Cycle-Consistency, Pragmatic-Load Disagreement, Manifold-Curvature Mismatch, and RIG. We argue that collapsing cultural competence into a single scalar incentivizes "probabilistic flattening," and that recognizing the unalignable class is a precondition for AI that respects, rather than erases, cultural divergence. This suggests that multilingual alignment is not a single well-defined objective, but a set of mutually incompatible projections.